Hypomethylation of the PXR gene promoter drives its overexpression in cisplatin-resistant EOC.

Through integrated analysis of gene expression profiles from 379 ovarian cancer samples in TCGA and 88 normal ovarian tissues in GTEx, we identified significant upregulation of PXR in malignant versus normal tissues (Fig.  1 A). Subsequent validation using the GEO dataset ( GSE160626 ) demonstrated markedly elevated PXR expression in cisplatin-resistant tumors compared to their sensitive counterparts (Fig.  1 B), suggesting a potential role for PXR in chemoresistance. Fig. 1 PXR drives ovarian cancer progression and cisplatin resistance. A Expression of PXR in ovarian cancer tissues and normal ovarian tissues. B PXR expression in cisplatin-resistant and cisplatin-sensitive groups of ovarian cancer. C The black curve represents the overall survival curve of the PXR low expression group, and the red curve represents the overall survival curve of the PXR high expression group. D Expression of PXR and MDR1 in ovarian cancer tissues. (* p  < 0.05) PXR drives ovarian cancer progression and cisplatin resistance. A Expression of PXR in ovarian cancer tissues and normal ovarian tissues. B PXR expression in cisplatin-resistant and cisplatin-sensitive groups of ovarian cancer. C The black curve represents the overall survival curve of the PXR low expression group, and the red curve represents the overall survival curve of the PXR high expression group. D Expression of PXR and MDR1 in ovarian cancer tissues. (* p  < 0.05) Kaplan-Meier survival analysis revealed significantly worse overall survival in patients with high PXR expression (Fig.  1 C), establishing its prognostic value. Furthermore, GEPIA database analysis revealed a significant positive correlation between PXR and its downstream target MDR1 in ovarian cancer specimens (Fig.  1 D), suggesting a plausible mechanistic link. Given the significant upregulation of PXR in malignant tissues, we evaluated its clinical relevance. Immunohistochemical analysis demonstrated progressively increasing PXR positivity from normal ovarian tissues to benign tumors, borderline tumors, and epithelial ovarian carcinomas (Fig.  2 A–D). Fig. 2 PXR overexpression associates with ovarian carcinogenesis and chemotherapy-induced adaptation. A Expression of PXR in different ovarian tissues. B Comparison of PXR protein immunohistochemical expression in epithelial ovarian cancer tissues and benign ovarian tumor tissues. C Comparison of PXR protein immunohistochemical expression in epithelial ovarian cancer tissues and normal ovarian tissues. D Comparison of PXR protein immunohistochemical expression in epithelial ovarian cancer tissues and ovarian contiguous tumor tissues. E Expression of PXR in surgically matched EOC tissues of the same patient before and after two chemotherapy surgeries. F Quantitative analysis of PXR expression in surgically matched EOC tissues of the same patient before and after two chemotherapy surgeries. G Expression of PXR in EOC tissues of chemotherapy-resistant and chemotherapy-sensitive groups. H Quantitative analysis of PXR expression in EOC tissues of chemotherapy-resistant and chemotherapy-sensitive groups. (* p  < 0.05; ** p  < 0.01; *** p  < 0.001) PXR overexpression associates with ovarian carcinogenesis and chemotherapy-induced adaptation. A Expression of PXR in different ovarian tissues. B Comparison of PXR protein immunohistochemical expression in epithelial ovarian cancer tissues and benign ovarian tumor tissues. C Comparison of PXR protein immunohistochemical expression in epithelial ovarian cancer tissues and normal ovarian tissues. D Comparison of PXR protein immunohistochemical expression in epithelial ovarian cancer tissues and ovarian contiguous tumor tissues. E Expression of PXR in surgically matched EOC tissues of the same patient before and after two chemotherapy surgeries. F Quantitative analysis of PXR expression in surgically matched EOC tissues of the same patient before and after two chemotherapy surgeries. G Expression of PXR in EOC tissues of chemotherapy-resistant and chemotherapy-sensitive groups. H Quantitative analysis of PXR expression in EOC tissues of chemotherapy-resistant and chemotherapy-sensitive groups. (* p  < 0.05; ** p  < 0.01; *** p  < 0.001) Analysis of clinicopathological parameters revealed no significant associations between PXR positivity and FIGO stage, ascites status, lymph node metastasis, or patient age. However, quantitative expression levels showed significant variation across FIGO stages (Tables  1 and 2 ). Table 1 Association between PXR positivity and clinicopathological characteristics in EOC Clinicopathological parameters Total cases ( n ) PXR expression p -value Positive (+) Negative (-) Positive rate (%) FIGO stage Stage I 8 8 0 100 0.131 d Stage II 10 9 1 90 Stage III 49 47 2 95.7 Stage IV 8 6 2 75 Ascites status Present 46 43 3 93.5 1.000 d Absent 29 27 2 93.1 Lymph node metastasis Positive 27 25 2 92.6 1.000 d Absent 33 30 3 90.9 Age (years) ≥ 54 39 35 4 89.7 0.360 d < 54 36 35 1 97.2 d: Fisher’s precision test Association between PXR positivity and clinicopathological characteristics in EOC d: Fisher’s precision test Table 2 Association between PXR expression levels and clinicopathological characteristics in EOC Clinicopathological parameters Total cases ( n ) PXR expression p -value Negative (-) Low (+) Moderate (++) High (+++) FIGO stage Stage I 8 0 3 2 3 Stage II 10 1 2 6 1 0.103 d Stage III 49 2 20 21 6 0.041 f Stage IV 8 2 5 1 0 Ascites status Present 46 3 16 19 8 0.522 d Absent 29 2 14 11 2 0.189 c Lymph node metastasis Positive 27 2 14 9 2 0.365 d Absent 33 3 11 12 7 0.181 c Age (years) ≥ 54 39 4 16 14 5 0.650 d < 54 36 1 14 16 5 0.348 c c: Rank sum test, d: Fisher’s exact test, f: H test Association between PXR expression levels and clinicopathological characteristics in EOC c: Rank sum test, d: Fisher’s exact test, f: H test Analysis of paired tumor tissues from 19 patients with epithelial ovarian cancer (EOC) before and after chemotherapy revealed significant differences in the distribution of PXR expression. Specifically, PXR expression in post-chemotherapy surgical lesions exhibited a polarized pattern: low expression accounted for 26.3%, while high expression increased to 47.4% (Fig.  2 E and F). In contrast, primary surgical lesions were predominantly characterized by low expression, followed by moderate expression. Notably, quantitative analysis further confirmed that the average PXR expression level in post-chemotherapy lesions was significantly higher than in primary lesions (Fig.  2 G and H). These findings suggest that chemotherapy may influence tumor biology by modulating PXR expression. Collectively, these findings implicate PXR in chemotherapy adaptation, showing progressive overexpression during malignant transformation and significant induction following therapeutic stress. PXR, a member of the nuclear receptor superfamily, directly regulates P-gp expression [ 27 , 28 ]. We investigated P-gp’s clinical relevance following established links between elevated PXR expression and poor ovarian cancer prognosis. Immunohistochemical analysis demonstrated a malignancy-associated P-gp expression gradient (Fig. 3 A), with positivity rates progressively increasing from normal ovarian tissues to benign tumors, borderline tumors, and epithelial ovarian carcinomas (Fig. 3 B–D). Significantly higher expression was confirmed in EOC versus benign tumors and normal tissues, supported by elevated expression intensity and distinct distribution patterns. Fig. 3 Malignancy-dependent P-gp overexpression and chemotherapy-responsive induction in ovarian carcinoma. A Expression of P-gp in different ovarian tissues. B Comparison of P-gp protein immunohistochemical expression in epithelial ovarian cancer tissues and benign ovarian tumor tissues. C Comparison of P-gp protein immunohistochemical expression in epithelial ovarian cancer tissues and normal ovarian tissues. D Comparison of P-gp protein immunohistochemical expression in epithelial ovarian cancer tissues and ovarian contiguous tumor tissues. E Expression of P-gp in surgically matched EOC tissues of the same patient before and after two chemotherapy surgeries. F Quantitative analysis of P-gp expression in surgically matched EOC tissues of the same patient before and after two chemotherapy surgeries. G Expression of P-gp in EOC tissues of chemotherapy-resistant and chemotherapy-sensitive groups. H Quantitative analysis of P-gp expression in EOC tissues of chemotherapy-resistant and chemotherapy-sensitive groups. (* p  < 0.05; ** p  < 0.01; *** p  < 0.001) Malignancy-dependent P-gp overexpression and chemotherapy-responsive induction in ovarian carcinoma. A Expression of P-gp in different ovarian tissues. B Comparison of P-gp protein immunohistochemical expression in epithelial ovarian cancer tissues and benign ovarian tumor tissues. C Comparison of P-gp protein immunohistochemical expression in epithelial ovarian cancer tissues and normal ovarian tissues. D Comparison of P-gp protein immunohistochemical expression in epithelial ovarian cancer tissues and ovarian contiguous tumor tissues. E Expression of P-gp in surgically matched EOC tissues of the same patient before and after two chemotherapy surgeries. F Quantitative analysis of P-gp expression in surgically matched EOC tissues of the same patient before and after two chemotherapy surgeries. G Expression of P-gp in EOC tissues of chemotherapy-resistant and chemotherapy-sensitive groups. H Quantitative analysis of P-gp expression in EOC tissues of chemotherapy-resistant and chemotherapy-sensitive groups. (* p  < 0.05; ** p  < 0.01; *** p  < 0.001) Notably, chemotherapy-induced dynamic P-gp modulation was observed in matched pre- and post-chemotherapy EOC tissues, which exhibited post-treatment shifts toward intermediate expression categories with elevated mean expression levels (Fig.  3 E and F). Furthermore, chemoresistant tissues demonstrated higher expression intensity and quantitative levels compared to chemosensitive counterparts (Fig.  3 G , H). No significant associations were observed between P-gp positivity and FIGO stage, ascites status, lymph node metastasis, or patient age. Collectively, these findings establish P-gp as both a marker of malignancy progression and a mediator of chemotherapy adaptation, operating independently of conventional clinicopathological parameters (Tables  3 and 4 ). Table 3 Relationship between the positive expression rate of P-gp and clinical pathological factors of OEC Clinicopathological parameters Total cases ( n ) P -gp expression χ2 p -value Positive (+) Negative (-) Positive rate (%) FIGO stage Stage I 8 8 0 100 0.131 d Stage II 10 9 1 90 Stage III 49 47 2 95.7 Stage IV 8 6 2 75 Ascites status Present 46 43 3 93.5 0.090 1.000 d Absent 29 27 2 93.1 Lymph node metastasis Positive 27 25 2 92.6 2.986 1.000 d Absent 33 30 3 90.9 Age (years) ≥ 54 39 35 4 89.7 0.028 0.360 d < 54 36 35 1 97.2 d: Fisher’s precision test Relationship between the positive expression rate of P-gp and clinical pathological factors of OEC d: Fisher’s precision test Table 4 The relationship between P-gp expression and EOC clinical and pathological factors Clinicopathological parameters Total cases ( n ) P -gp expression p -value Negative (-) Low (+) Moderate (++) High (+++) FIGO stage Stage I 8 2 5 1 0 Stage II 10 4 5 1 0 0.935 d Stage III 49 22 22 4 1 0.803 f Stage IV 8 4 3 1 0 Ascites status Present 46 19 21 6 0 0.355 d Absent 29 13 14 1 1 0.618 c Lymph node metastasis Positive 27 15 10 2 0 0.229 d Absent 33 11 18 4 0 0.097 c Age (years) ≥ 54 39 17 20 2 0 0.373 d < 54 36 15 15 4 1 0.486 c c: Rank sum test, d: Fisher’s exact test, f: H test The relationship between P-gp expression and EOC clinical and pathological factors c: Rank sum test, d: Fisher’s exact test, f: H test PXR and P-gp exhibit synergistic upregulation during ovarian malignant transformation and chemotherapy adaptation, with their co-overexpression significantly correlating with therapeutic resistance, suggesting that this regulatory axis is a potential target for overcoming chemoresistance in EOC. Integrated bioinformatics analysis coupled with immunohistochemical staining revealed a significant positive correlation between PXR and P-gp expression in ovarian cancer specimens. This molecular association prompted further investigation into their coordinated regulation. Quantitative RT-PCR analysis confirmed detectable mRNA expression of both PXR and MDR1 (encoding P-gp) across all tested cell lines (SKOV3, OVCAR3, A2780, A2780/DDP). Notably, cisplatin-resistant A2780/DDP cells exhibited significantly elevated mRNA levels compared to parental lines (SKOV3/OVCAR3/A2780), demonstrating a stepwise increase (Fig.  4 A, B). Western blot analysis validated the concordant upregulation of PXR and P-gp proteins in chemoresistant cells, with GAPDH used as a loading reference (Fig.  4 C). Fig. 4 PXR-MDR1 transcriptional axis activation mediates intrinsic and acquired chemoresistance in ovarian carcinoma. A qRT-PCR was used to detect the expression of PXR mRNA in ovarian cancer cell lines SKOV3, OVCAR3, A2780 and A2780/DDP. B qRT-PCR was used to detect the expression of MDR1 mRNA in ovarian cancer cell lines SKOV3, OVCAR3, A2780 and A2780/DDP. C Western blot was used to detect the expression of PXR and P-gp in cells. D qRT-PCR was used to detect the knockdown of PXR in A2780/DDP transfected cells. E Western blot was used to detect the expression of PXR protein in cells after PXR knockdown. (* p  < 0.05; ** p  < 0.01; *** p  < 0.001) PXR-MDR1 transcriptional axis activation mediates intrinsic and acquired chemoresistance in ovarian carcinoma. A qRT-PCR was used to detect the expression of PXR mRNA in ovarian cancer cell lines SKOV3, OVCAR3, A2780 and A2780/DDP. B qRT-PCR was used to detect the expression of MDR1 mRNA in ovarian cancer cell lines SKOV3, OVCAR3, A2780 and A2780/DDP. C Western blot was used to detect the expression of PXR and P-gp in cells. D qRT-PCR was used to detect the knockdown of PXR in A2780/DDP transfected cells. E Western blot was used to detect the expression of PXR protein in cells after PXR knockdown. (* p  < 0.05; ** p  < 0.01; *** p  < 0.001) Given maximal PXR expression in A2780 and A2780/DDP cells, targeted siRNA silencing was performed. qRT-PCR was used to verify the level of regulation. qRT-PCR showed that PXR mRNA levels in siNC-transfected A2780/DDP cells were not significantly different from those in untransfected A2780/DDP control cells. Compared with the control siNC group, the expression level of PXR in the siPXR group was reduced considerably, and the siPXR-3 group showed the highest knockdown efficiency (Fig.  4 D). Western blot results were consistent with qRT-PCR (Fig.  4 E), so we chose siPXR-3 to construct an ovarian cancer cell model with knockdown of the PXR gene. The regulated cells were used in subsequent experiments examining cell migration, invasion, proliferation, and cisplatin resistance. To elucidate the upstream epigenetic mechanisms underlying aberrant PXR expression in cisplatin-resistant EOC, we first compared genome-wide DNA methylation levels in cisplatin-sensitive (A2780) and cisplatin-resistant (A2780/DDP) cells. Results showed a significant decrease in global 5mC methylation in A2780/DDP cells (Fig.  5 A), suggesting the activation of active demethylation in resistant cells. Further analysis revealed that TET1, a key enzyme responsible for catalyzing the conversion of 5mC to 5hmC, was significantly upregulated in resistant cells (Fig.  5 B and C), suggesting that it may be an upstream factor driving aberrant PXR activation. Fig. 5 TET1 regulates PXR expression through promoter demethylation. A Global 5mC methylation levels in cisplatin-sensitive A2780 and cisplatin-resistant A2780/DDP cells; B , C qPCR and Western blot analysis examined TET1 expression in A2780 cells and A2780/DDP cells; D , E Effects of TET1 knockdown on PXR mRNA and protein expression in A2780/DDP cells; F ChIP-qPCR demonstrating TET1 binding to the PXR promoter and enrichment of 5hmC in A2780/DDP cells, both of which were diminished upon TET1 knockdown; G Detection of apoptosis in A2780/DDP cells following TET1 knockdown. (** p  < 0.01; *** p  < 0.001) TET1 regulates PXR expression through promoter demethylation. A Global 5mC methylation levels in cisplatin-sensitive A2780 and cisplatin-resistant A2780/DDP cells; B , C qPCR and Western blot analysis examined TET1 expression in A2780 cells and A2780/DDP cells; D , E Effects of TET1 knockdown on PXR mRNA and protein expression in A2780/DDP cells; F ChIP-qPCR demonstrating TET1 binding to the PXR promoter and enrichment of 5hmC in A2780/DDP cells, both of which were diminished upon TET1 knockdown; G Detection of apoptosis in A2780/DDP cells following TET1 knockdown. (** p  < 0.01; *** p  < 0.001) We knocked down TET1 in A2780/DDP cells. Results revealed a significant decrease in both PXR mRNA and protein levels (Fig.  5 D and E), demonstrating that TET1 is required for maintaining elevated PXR expression. We then examined the methylation status of the PXR promoter region using Chip-qPCR and found that TET1 specifically binds to this region and significantly enriches 5hmC. Both binding and enrichment were significantly reduced upon TET1 knockdown (Fig.  5 F). These results suggest that TET1 catalyzes the generation of 5hmC at the PXR promoter, thereby promoting its demethylation, relieving transcriptional repression and driving elevated PXR expression. Finally, functional rescue experiments showed that TET1 knockdown significantly enhanced the sensitivity of A2780/DDP cells to cisplatin (Fig.  5 G), further demonstrating that PXR is a key downstream effector molecule of TET1, mediating TET1-related cisplatin resistance. To elucidate the role of PXR in metastatic progression, we systematically assessed the migration and invasion abilities of cells following PXR knockdown. Scratch wound healing assays demonstrated that PXR knockdown significantly impaired cell migration: in both A2780/DDP and A2780 cells, wound closure in the siPXR group was significantly delayed compared to the negative control group (Figs.  6 A–D), with statistically significant differences. Fig. 6 PXR knockdown suppresses metastatic potential. A Scratch assay to detect changes in the migration ability of A2780 cells after knockdown of PXR. B Changes in migration ability of A2780/DDP cells after knockdown of PXR; C Statistical graph of the scratch assay of A2780 cells. D Statistical graph of the scratch assay of A2780/DDP cells. E Transwell assay to detect changes in the invasion ability of A2780 cells after knockdown of PXR. F Transwell assay to detect changes in the invasion ability of A2780/DDP cells after knockdown of PXR (scale bar is 50 μm). G Statistical graph of the Transwell assay of A2780 cells. H Statistical graph of the Transwell assay of A2780/DDP cells. (** p  < 0.01; *** p  < 0.001) PXR knockdown suppresses metastatic potential. A Scratch assay to detect changes in the migration ability of A2780 cells after knockdown of PXR. B Changes in migration ability of A2780/DDP cells after knockdown of PXR; C Statistical graph of the scratch assay of A2780 cells. D Statistical graph of the scratch assay of A2780/DDP cells. E Transwell assay to detect changes in the invasion ability of A2780 cells after knockdown of PXR. F Transwell assay to detect changes in the invasion ability of A2780/DDP cells after knockdown of PXR (scale bar is 50 μm). G Statistical graph of the Transwell assay of A2780 cells. H Statistical graph of the Transwell assay of A2780/DDP cells. (** p  < 0.01; *** p  < 0.001) Transwell invasion assays further validated these findings: siPXR treatment significantly reduced the number of invasive cells in both cell lines compared to the control group (Figs.  6 E–H), with statistically significant differences. This synergistic reduction in migration and invasion abilities suggests that PXR is a key regulator of metastatic spread in chemoresistant epithelial ovarian cancer (EOC). We further investigated the functional impact of PXR on tumor growth dynamics and therapeutic resistance. CCK-8 assays revealed attenuated proliferative capacity upon PXR knockdown. Both A2780/DDP and A2780 cells showed significantly lower viability compared with siNC controls (Fig.  7 A and B). Fig. 7 PXR regulates proliferation and chemoresistance. A Survival rate of A2780/DDP cells transfected with siNC and siPXR detected by CCK-8; B Survival rate of A2780 cells transfected with siNC and siPXR detected by CCK-8; C Survival rate of A2780/DDP cells transfected with siPXR and siNC detected by CCK-8 and cisplatin treatment; D Statistical graph of IC50 values of A2780/DDP cells transfected with siPXR and siNC and cisplatin treatment. (* p  < 0.05; ** p  < 0.01) PXR regulates proliferation and chemoresistance. A Survival rate of A2780/DDP cells transfected with siNC and siPXR detected by CCK-8; B Survival rate of A2780 cells transfected with siNC and siPXR detected by CCK-8; C Survival rate of A2780/DDP cells transfected with siPXR and siNC detected by CCK-8 and cisplatin treatment; D Statistical graph of IC50 values of A2780/DDP cells transfected with siPXR and siNC and cisplatin treatment. (* p  < 0.05; ** p  < 0.01) Cisplatin sensitivity assays demonstrated restored drug responsiveness in PXR-deficient cells. siPXR-transfected A2780/DDP cells exhibited reduced IC50 values and markedly lower survival across cisplatin concentrations (Fig.  7 C and D). The dual suppression of proliferative capacity and chemoresistance positions PXR as a pivotal therapeutic node for EOC treatment optimization.

Normal ovarian tissue RNA-seq data ( n  = 88) were obtained from GTEx (v8), while ovarian cancer transcriptomes ( n  = 379) were acquired from the TCGA-OV project. Additional validation datasets ( GSE160626 ) containing eight platinum-sensitive and 10 platinum-resistant specimens were retrieved from GEO. All datasets were processed through a standardized pipeline: (1) raw counts were normalized using DESeq2 variance-stabilizing transformation; (2) batch effects were corrected via ComBat-seq; (3) gene expression was quantified as log2(TPM + 1). Survival analysis was conducted using Kaplan-Meier Plotter with optimal expression cutoff determination (maximally selected rank statistics). The association between PXR expression (high vs. low) and overall survival was assessed using the log-rank test (significance threshold: p  < 0.05). For co-expression analysis, GEPIA2 was employed to calculate Pearson correlation coefficients between PXR (NR1I2) and MDR1 (ABCB1) using TCGA-OV data, with significance determined by two-tailed t-test. All statistical analyses and visualizations were performed using GraphPad Prism 9 (unpaired t-test for group comparisons, α = 0.05). This study was approved by the Ethics Committee of Yichang Central People’s Hospital (Approval No. 2022-138-01). A total of 194 paraffin-embedded ovarian tissue specimens and corresponding clinical data were collected from patients who underwent surgical treatment at the Department of Gynecology, The First Clinical Medical College of Three Gorges University between January 2017 and June 2022. The specimens included 94 EOC samples from 75 patients (including 38 paired specimens from 19 patients who underwent both primary and secondary surgeries after chemotherapy), 40 ovarian borderline tumors, 40 benign ovarian tumors, and 20 normal ovarian tissues (obtained during hysterectomy for uterine fibroids or adenomyosis with pathologically confirmed standard ovarian structure). The EOC group consisted of 75 treatment-naïve patients. Control groups included borderline tumors, benign tumors, and normal ovaries, with no significant age differences among the groups ( p  > 0.05). Inclusion criteria required: (1) primary surgical treatment with pathological confirmation at our institution, (2) complete clinical/pathological records and follow-up data, (3) exclusion of metastatic cancers, non-primary ovarian malignancies, immune disorders, or concurrent other organ tumors. All EOC patients received platinum-based adjuvant chemotherapy. Comprehensive clinical parameters were collected for all 75 EOC patients, including age, FIGO stage (according to the 2014 criteria), histological type, tumor differentiation, presence of ascites, lymph node metastasis status, chemotherapy responsiveness (sensitive or resistant), and menopausal status. The cohort was stratified by age and FIGO stage. Histological analysis revealed 94.7% serous carcinomas, 4.0% clear cell carcinomas, and 1.3% endometrioid carcinomas. Ascites was present in 61.3% of cases. All patients underwent primary surgery without prior chemotherapy or radiotherapy. Surgical approaches varied by stage: early-stage (I-II) patients received hysterectomy with unilateral/bilateral salpingo-oophorectomy ± omentectomy/appendectomy/pelvic lymphadenectomy. At the same time, advanced cases (III-IV) underwent tumor debulking with comprehensive resection. Lymphadenectomy was performed in 80.0%, revealing nodal metastases in 45.0%. Postoperative platinum-based chemotherapy was administered to 96.0% of patients, with 25.0% developing chemoresistance and 75.0% showing sensitivity. The clinicopathologic data of the patients are shown in Table S1. All EOC cases underwent comprehensive follow-up through outpatient reviews (including physical examinations, serum tumor marker tests, and imaging studies) and telephone interviews to document adjuvant treatment responses, survival status, mortality dates, and causes of death. The standardized follow-up protocol mandated quarterly evaluations for the first two postoperative years, followed by biannual to annual assessments thereafter. The study’s follow-up period spanned from the surgery dates until June 2023 (with a duration of 12–72 months), and overall survival was calculated in months from surgery to either death or the censoring date. This rigorous surveillance ensured the collection of complete clinicopathological data for all enrolled patients. Immunohistochemical staining was performed on four µm-thick formalin-fixed paraffin-embedded tissue sections. The slides were first baked at 75 °C for 40 min to enhance tissue adhesion. Deparaffinization was achieved through sequential immersion in xylene and graded ethanol solutions, followed by rehydration in distilled water. For antigen retrieval, slides were treated with EDTA solution using a pressure cooker for 25 min. Endogenous peroxidase activity was blocked by incubating sections in 3% hydrogen peroxide solution for 10 to 15 min at room temperature. Primary antibody incubation was performed at 37 °C for 60 min in an automated stainer, followed by the application of HRP-conjugated secondary antibody under identical temperature conditions for 30 min. Chromogenic detection was performed using freshly prepared DAB solution for precisely 7 min, after which the reaction was stopped by thorough washing. Nuclear counterstaining was achieved with hematoxylin to provide appropriate contrast. Finally, slides were dehydrated through an ethanol series, cleared in xylene, and permanently mounted for microscopic evaluation. Quantitative image analysis was conducted using Image Pro Plus 7.0 software to ensure objective assessment of staining patterns. SKOV3 was cultured in DMEMF12 medium. OVCAR3, A2780, and A2780/DDP were maintained in RPMI-1640 Medium supplemented with 10% fetal bovine serum, 100 U/mL penicillin, and 100 µg/mL streptomycin. The cisplatin-resistant A2780/DDP cell line was generated by continuous exposure of parental A2780 cells to increasing concentrations of cisplatin, as previously described [ 26 ]. The cells were cultured in a humidified incubator at 37 °C with 5% CO2. To knock down the expression of PXR, siRNAs targeting PXR were utilized. The sequences of these siRNAs are provided in Table S2. Transfection was conducted using Lipofectamine 3000 according to the manufacturer’s instructions. Briefly, cells were seeded into 6-well plates and cultured in a humidified incubator at 37 °C with 5% CO2. Once the cells reached 70–80% confluence, they were transfected with 50 nM of either PXR siRNA using Lipofectamine 3000 and subsequently cultured for an additional four hours. After the siRNA treatment, the cells were collected by centrifugation at 500 g for 15 min at 4 °C. The siRNA sequence information used for PXR gene specific knockdown is detailed in Supplementary Material Table S2. Cellular proteins were extracted using RIPA lysis buffer (AbMole BioScience) supplemented with protease inhibitors. Protein lysates were resolved by 10% SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and electrophoretically transferred onto 0.5 μm polyvinylidene difluoride (PVDF) membranes. Membranes were blocked with 5% skim milk in Tris-buffered saline with Tween-20 (TBST) for one hour at 25 °C, followed by three 5-minute washes with TBST. Subsequently, the membranes were incubated overnight at 4 °C with primary antibodies, washed three times with TBST (10 min per wash), and probed with horseradish peroxidase (HRP)-conjugated secondary antibodies (1:10,000 dilution) for 1 h at 25 °C. After final TBST washes (3 × 10 min), immunoreactive bands were visualized using the ChemiDoc Touch Imaging System (Bio-Rad) with enhanced chemiluminescence substrate. β-Actin served as the loading control for protein normalization. Total RNA was extracted using TRIzol (Beyotime). Approximately 2 × 10 6 cells were resuspended in 1 mL of TRIzol, followed by the addition of 0.1 mL of chloroform to denature proteins. After centrifugation at 10,000 g for 15 min at 4 °C, the supernatant was collected, and RNA was purified through isopropyl alcohol precipitation. Reverse transcription of the RNA was conducted using the HiScript III RT SuperMix (Biosharp, Hefei, China). Subsequently, real-time quantitative PCR (qPCR) was performed with a CFX Connect Real-Time PCR Detection System (Bio-Rad) using SYBR Green Master Mix (Takara Bio). Each qPCR reaction mixture (10 µL) contained 5 µL of 2× SYBR Green Master Mix, 0.5 µL of forward primer (10 µM), 0.5 µL of reverse primer (10 µM), 2 µL of cDNA template, and 2 µL of RNase-free water. The thermal cycling protocol consisted of an initial denaturation step at 95 °C for 30 s, followed by 40 cycles of denaturation at 95 °C for 5 s, annealing at 53–57 °C (depending on the specific primers used) for 30 s, and extension at 72 °C for 30 s. Details of the primer sequences are provided in Table S3. Cells were seeded in 6-well plates at a density of 5 × 10^5 cells per well and cultured to 90–100% confluence. A standardized wound field was created by scratching the monolayer with a sterile 200 µL pipette tip, followed by three PBS washes to remove detached cells. Wound closure was monitored at 0 and 48 h under phase-contrast microscopy (Nikon Eclipse Ti2, 10× objective), with images captured at three predetermined positions per well using NIS-Elements software. Migration distance was quantified by measuring the residual wound width (ImageJ v1.53) and expressed as percentage closure relative to the initial wound area (t0). Three independent experiments were performed in triplicate, with data normalized to control groups. Statistical analysis was conducted using GraphPad Prism 9.0 (two-way ANOVA with Tukey’s post-hoc test). Cell viability was quantitatively assessed using the CCK-8 assay (Dojindo, Japan) according to the manufacturer’s protocol with optimizations. Briefly, cells were seeded in 96-well plates at a density of 3 × 10³ cells/well and incubated for 24 h to allow for attachment. After treatment with specified compounds or controls, 10 µL of CCK-8 reagent was added to each well, followed by incubation at 37 °C for two hours. Absorbance was measured at 450 nm using a microplate reader (BioTek Synergy H1) with 650 nm as reference wavelength. Six replicate wells were included per condition, with blank correction using medium-only wells. For IC50 determination, cells were treated with serial dilutions of cisplatin (0–100 µM) for 48 h before CCK-8 measurement. Dose-response curves were generated using nonlinear regression analysis (GraphPad Prism 9.0, four-parameter logistic model). Three independent experiments were performed, with data expressed as percentage viability relative to untreated controls. Cell invasion capacity was evaluated using Corning Matrigel-coated Transwell chambers (8-µm pore size, 24-well format) according to standardized protocols with modifications. Briefly, the upper chambers were pre-coated with 50 µL of growth factor-reduced Matrigel (diluted 1:8 in serum-free medium) and polymerized at 37 °C for 4 h. Serum-starved cells (5 × 10^4 in 200 µL of serum-free medium) were then seeded into the upper chamber. In comparison, the lower chamber contained 600 µL complete medium with 10% FBS as a chemoattractant. After 24 h incubation (37 °C, 5% CO₂), non-invaded cells on the upper membrane surface were removed with cotton swabs. The invaded cells on the lower surface were fixed with 4% paraformaldehyde (15 min) and stained with 0.1% crystal violet (20 min). Five random fields per insert were imaged under phase-contrast microscopy (Nikon Eclipse Ti, 20× objective), with cell counts performed using ImageJ v1.53 automated particle analysis. Three independent experiments were conducted in triplicate, with data normalized to the control group invasion (set as 100%). Statistical significance was determined by one-way ANOVA with Tukey’s post-hoc test (GraphPad Prism 9.0).

This study establishes that the pregnane X receptor (PXR) is a key driver of epithelial ovarian cancer progression and chemoresistance through transcriptional regulation of MDR1/P-gp. Through integrated analyses of clinical specimens and experimental models, we found that PXR expression increases progressively during ovarian malignant transformation and peaks in chemoresistant tumors. This clinical pattern correlates with poor patient prognosis, suggesting that PXR serves as both a biomarker for aggressive disease and a functional mediator of therapeutic resistance. We demonstrate that the aberrantly elevated PXR expression in chemoresistance is driven upstream by an active epigenetic mechanism. We found that the active demethylase TET1 is upregulated in resistant cells and catalyzes the generation of 5-hydroxymethylcytosine in the PXR gene promoter, leading to hypomethylation and thus deregulating PXR transcription. The significant induction of PXR expression after chemotherapy exposure, particularly in resistant cases, suggests a role in adaptation to therapeutic stress, and our findings provide an epigenetic explanation for this adaptation. These findings were further strengthened by the coordinated upregulation of PXR and its downstream target MDR1 [ 29 ], forming a functional resistance axis independent of traditional clinicopathological parameters. At the molecular level, we identified the PXR-MDR1 axis as a core mechanism of cisplatin resistance in ovarian cancer. This newly elucidated signaling pathway, TET1-PXR-MDR1, showed strong associations in both clinical specimens and cell line models. Functional validation by PXR knockdown not only reduced MDR1 expression but also restored drug sensitivity in resistant cells, confirming the causal role of this axis in chemotherapy resistance. Although MDR1/P-glycoprotein is the most widely recognized downstream effector of PXR, cisplatin resistance is now widely acknowledged as a multifactorial process involving the coordinated dysregulation of multiple ATP-binding cassette (ABC) transporters. Beyond MDR1, transporters such as ABCC1 (MRP1) and ABCG2 (BCRP) also contribute to platinum resistance by limiting intracellular drug accumulation. Given PXR’s role as a broad-spectrum transcriptional regulator of heterogenous transporters and metabolic enzymes, its overexpression is more likely to function as an upstream regulatory hub coordinating broader resistance networks rather than acting solely through MDR1. In addition to affecting drug efflux, PXR appears to maintain tumor cell viability under therapeutic stress, as demonstrated by experiments showing that PXR-deficient cells had reduced proliferation even in the absence of cisplatin. This suggests that PXR coordinates a broader range of survival pathways, with functions beyond its classical role in drug metabolism [ 30 ]. Our studies also reveal a previously unrecognized role for PXR in the metastatic progression of ovarian cancer. The significant impairment of cell migration and invasion following PXR knockdown positions this nuclear receptor as a key regulator of metastatic behavior. Given that TET1-mediated active demethylation is often associated with rapid gene expression reprogramming and cellular plasticity, we hypothesize that the TET1-PXR axis may play a broader role in regulating genes involved in epithelial-mesenchymal transition or cytoskeletal remodeling, thereby driving both drug resistance and metastasis. This finding provides a potential explanation for the poor prognosis observed in patients with high PXR expression, independent of traditional metastatic markers. This study has some limitations that need to be addressed. First, although we integrated bioinformatics analysis, clinical samples, and in vitro functional experiments, the mechanistic findings are primarily based on cell models and have not been validated in vivo. Future studies should utilize animal models to further confirm the functional role of the TET1-PXR axis in cisplatin resistance. Second, the cisplatin resistance phenotype was mainly studied using the A2780/DDP cell line; using other resistance models would help enhance the generalizability of our conclusions. These collective findings have important implications for the therapeutic management of epithelial ovarian cancer. From initial carcinogenesis to acquired chemoresistance, the TET1-PXR-MDR1 axis is consistently upregulated across various disease states, highlighting its fundamental role in disease progression and establishing PXR and its upstream regulator, TET1, as promising therapeutic targets. Although PXR antagonists developed for metabolic diseases have the potential to be repurposed, the ubiquitous expression of PXR in normal tissues may require tumor-specific delivery methods to minimize toxicity [ 31 ]. Our observation that chemotherapy itself induces PXR upregulation suggests that the optimal treatment strategy may require the combined use of PXR or TET1 inhibitors after standard chemotherapy to prevent the emergence of drug resistance. Future studies should explore the specific downstream effectors that mediate the pro-metastatic effects of PXR and investigate combined strategies that simultaneously target the TET1-PXR axis and conventional chemotherapy pathways. In summary, this study systematically demonstrates a key role of PXR in cisplatin resistance in EOC and further deepens our understanding of the underlying resistance mechanisms by elucidating a TET1-mediated epigenetic regulatory pathway, thereby providing a potential target for the development of new therapeutic strategies. With further research on the role of PXR and its epigenetic regulation in tumors, therapeutic approaches targeting the TET1-PXR pathway are expected to bring new hope for improving the prognosis of EOC patients.

Epithelial ovarian cancer (EOC) is the most lethal gynecological malignancy [ 1 , 2 ], with approximately 90% of cases eventually developing platinum resistance [ 3 ], which has become the most critical treatment bottleneck for improving patients’ clinical prognosis [ 4 ]. Although cytoreductive surgery combined with platinum chemotherapy is the current standard treatment, the vast majority of patients inevitably face disease recurrence and drug resistance, leading to treatment failure [ 5 – 7 ]. This treatment dilemma highlights the urgency of deciphering the molecular mechanisms of EOC resistance, especially the paradoxical phenomenon that chemotherapy itself may induce adaptive resistance [ 8 , 9 ]. Therefore, a comprehensive understanding of the molecular pathways underlying ovarian cancer progression and resistance is crucial for advancing early diagnosis, overcoming treatment resistance, and improving patient prognosis [ 10 ]. Pregnane X receptor (PXR/NR1I2), a ligand-activated nuclear receptor, has functions far beyond its classical liver detoxification role and has been shown to be a core regulator of chemotherapy resistance [ 11 – 14 ]. Mechanistically, PXR can transcriptionally regulate multidrug resistance protein 1 (MDR1/P-glycoprotein), an ATP-binding cassette transporter that effluxes chemotherapeutic drugs such as platinum and paclitaxel [ 15 – 18 ]. Emerging evidence further reveals the broader oncogenic function of PXR: in colorectal cancer models, cancer stem cells drive relapse after treatment by overexpressing PXR [ 19 ]; and in a variety of aggressive cancers, high expression of PXR is closely associated with poor patient prognosis [ 20 , 21 ]. Although PXR plays an important role in the progression and chemotherapy resistance of various malignancies [ 22 , 23 ], its specific mechanism of action in ovarian cancer, especially the upstream events that lead to its own abnormal expression, remains unclear. It is worth noting that epigenetic regulation, especially DNA methylation, plays a decisive role in gene expression reprogramming and cancer pathogenesis. Promoter hypomethylation is a common mechanism of oncogene activation [ 24 ]. Based on this, we hypothesized that PXR overexpression in chemotherapy-resistant EOC may stem from DNA hypomethylation in its promoter region [ 25 ]. In this study, we aimed to systematically investigate the role of PXR in epithelial ovarian cancer progression and cisplatin resistance. Specifically, we examined the expression pattern and clinical relevance of PXR in ovarian cancer tissues, explored the epigenetic mechanisms regulating PXR transcription with a focus on TET1-mediated promoter hypomethylation, and evaluated the functional impact of the TET1–PXR–MDR1 axis on tumor cell proliferation, migration, invasion, and chemoresistance. By integrating bioinformatic analyses, clinical sample validation, and mechanistic experiments, this study seeks to provide a comprehensive framework for understanding epigenetically driven PXR activation and its contribution to therapeutic resistance in epithelial ovarian cancer.

Supplementary Material 1. Supplementary Material 1.