Gene expression analysis of DHCR24 in OC tissues versus normal ovarian epithelial tissues was performed using the Gene Expression Profiling Interactive Analysis 2 (GEPIA 2) (http://gepia.cancer-pku.cn), which integrates data from the Cancer Genome Atlas (TCGA) database. The analysis included 426 tumor samples and 88 normal control samples. The “Expression DIY” module in GEPIA 2 was employed for comparative analysis, with a statistical significance threshold set at p < 0.05. Prognostic analysis of DHCR24 expression was performed using the Kaplan-Meier Plotter (https://kmplot.com/analysis/, accessed on 14 May 2025) online platform with progression-free survival (PFS) as the primary endpoint. Patients were stratified into high- and low-expression groups based on the optimal expression cutoff value. Intergroup differences were assessed by log-rank test with a significance threshold of p < 0.05.

In this study, paraffin sections of 70 cases of OC were examined. In addition, fresh human normal ovarian and OC tissue specimens (16 each) were collected and stored in liquid nitrogen for follow-up experiments. Patient details can be found in Table S1.

This study was approved by the Ethics Committee of the First Affiliated Hospital of Chongqing Medical University (approval number: 2024-006-02). Informed consent has been obtained from all patients.

Normal ovarian epithelial cells IOSE-80 (RRID: CVCL_5546) were purchased from Zhong Qiao Xin Zhou Biotechnology Co., Ltd. (Shanghai, China). OC cells [SKOV3 (RRID:CVCL_0532), CAOV3 (RRID:CVCL_0201), ES2 (RRID: CVCL_3509), and OVCAR3 (RRID:CVCL_0465)]. OC cells A2780 (RRID: CVCL_0134) were purchased from Servicebio Co., Ltd. (Wuhan, China). All cell lines were authenticated by STR profiling and confirmed to be free of mycoplasma contamination. All cells were cultured in McCoy’s 5A medium (Procell, Wuhan, China) at 37°C in a 5% CO₂ incubator.

Total RNA was extracted from the aforementioned tissues and cells using Trizol (Cat. No. 9109, TaKaRa, Tokyo, Japan). qPCR was performed by TB Green Premix Ex Taq (Cat. No. RR820A, TaKaRa) [31,32]. Gene expression differences were calculated using the 2^−ΔΔCt method, with intergroup comparisons performed by t-tests or one-way ANOVA. Primer sequences are shown in Table S2.

Proteins (25 μg) were separated by 8% and 10% sodium dodecyl-sulfate polyacrylamide gel electrophoresis. The proteins were transferred onto a polyvinylidene difluoride membrane via electroblotting. The membrane was blocked with skim milk for 2 h, followed by incubation with primary antibodies [diluted in antibody dilution buffer (Cat. No. P0256, Beyotime, Shanghai, China)] at 4°C overnight, using GAPDH or β-Actin as controls [with the following dilution ratios: DHCR24 1:1000 (#2033, CST, Danvers, MA, USA), E-cadherin 1:1000 (#3195, CST, USA), N-cadherin 1:1000 (#13116, CST, USA), Vimentin 1:1000 (#5741, CST, USA), Bax 1:1000 (#5023, CST, USA), Bcl2 1:1000 (#3498, CST, USA), TGF-β1 1:1000 (ab215715, Abcam, Cambridge, UK), Smad2/3 1:1000 (#8685, CST, USA), Phospho-Smad2 1:1000 (#3108, CST, USA), Phospho-Smad3 1:2000 (ab52903, Abcam, UK), GAPDH 1:5000 (380626, Zen-BioScience, Chengdu, China), β-Actin 1:5000 (380624, Zen-BioScience, China)]. The next day, secondary antibody incubation (diluted in Tris-Buffered Saline with Tween-20) was performed at room temperature for 1 h. Finally, an ultrasensitive chemiluminescence kit (Cat. No. BL520A, Biosharp, Anhui, China) was used to visualize the proteins. Antibody data are presented in Table S3. All experiments were independently repeated three times.

The cells (SKOV3, CAOV3, ES2, and A2780) were seeded (1 × 10⁵ cells/well) in a 12-well plate one day in advance to achieve approximately 30% confluence by the second day. SKOV3 and CAOV3 cells were infected with a lentivirus-mediated DHCR24 low-expression vector (Hanbio, Shanghai, China). Resistant cells were screened using puromycin (2 µg/mL), resulting in stable DHCR24 low-expression cell lines SKOV3/sh1, sh2, sh3, and their control cell lines SKOV3/shNC, alongside CAOV3/sh1, sh2, sh3, and their control cell lines CAOV3/shNC. In addition, ES2 and A2780 cells were infected with a lentivirus-mediated DHCR24 overexpression vector (Hanbio). Stable DHCR24 high-expression cell line ES2/DHCR24 and its control cell line ES2/Vector, and A2780/DHCR24 and its control cell line A2780/Vector were constructed. DHCR24 interference sequences are shown in Table S4.

Cells (3 × 10³ cells/well) were evenly seeded in a 96-well plate, with five replicate wells per group, and blank control wells (containing only medium + CCK-8 reagent) were also set up. Five plates were prepared simultaneously for each group (labeled as D0–D4). After 4−6 h, cell adherence was checked, and the medium was replaced with a fresh medium containing 10% CCK-8 solution (Cat. No. K1018, APExBIO, Houston, TX, USA). The plates were then returned to the incubator for further culturing. After 2 h, the D0 plate was removed, and absorbance at 450 nm was measured under light-protected conditions using a microplate reader (Infinite 200 PRO, Tecan, Switzerland), recorded as the D0 absorbance value. The same procedure was repeated at the same time for D1–D4. First, measure the raw absorbance, then subtract the blank control to calculate the corrected absorbance: Corrected absorbance = Sample well absorbance − Blank well absorbance.

Cells (SKOV3, CAOV3, ES2 and A2780) were evenly seeded (4 × 10² cells/well) and cultured. Samples were collected once colonies visible to the naked eye had formed (approximately 2 weeks of culturing). Photographs were captured and the colony numbers were quantified using Fiji software (version 2.3.0).

For the migration assay, SKOV3 and ES2 cells (2 × 10⁴ cells/well) and CAOV3 and A2780 cells (6 × 10⁴ cells/well) were inoculated into Transwell inserts (BIOFIL, Guangzhou, China) containing fetal bovine serum-free medium, with the complete medium in the lower chamber. For the invasion assay, 80 μL of Matrigel (Cat. No. 082704, Mogengel-Bio, Xiamen, China) diluted at a 1:8 ratio was first added to the upper chamber and incubated for 3 h to allow the gel to solidify. The cells were then seeded (same as the migration assay). The upper and lower chambers were supplemented with the same medium as previously described. Samples were collected after 24 or 96 h of incubation. Specifically, after 24 h, samples were collected from SKOV3 and ES2 cells, whereas after 96 h, samples were collected from CAOV3 and A2780 cells. Samples were first fixed with 4% paraformaldehyde solution (Cat. No. P1110, Solarbio, Beijing, China) for 15 min, then stained with crystal violet solution (Cat. No. C0121, Beyotime, Shanghai, China) for 9–10 min, washed with PBS, and subsequently photographed.

Cells (SKOV3 and CAOV3) were seeded (5 × 10⁵ cells/well) in a 6-well plate and cultured for approximately 3−4 days. When the confluence reached over 90%, a 200 µL pipette tip was used to create linear scratches perpendicular to the bottom of the plate. A photomicrograph of the scratched area was taken, marking the time as 0 h. The culture was continued, and images of the scratch area were captured at appropriate time points (after 24 h for SKOV3 cells; after 72 h for CAOV3 cells) using a microscope (CKX53, OLYMPUS, Tokyo, Japan).

Cells (SKOV3, CAOV3, ES2, and A2780) were seeded (5 × 10⁵ cells/well) and sampled as the cells grew logarithmically. Binding buffer, annexin V-FITC, and propidium iodide (Cat. No. E-CK-A211, Elabscience, Wuhan, China) were then slowly added. Incubate the cells for 15 min in the dark before analysis. The flow cytometer used in this experiment was the CYTOFLEX (CytoFLEX S, Beckman Coulter, Brea, CA, USA).

Twenty-two BALB/c nude mice were purchased from Changzhou Cavins Laboratory Animal Co., Ltd. (Changzhou, China). The mice were housed in an AAALAC-accredited facility with controlled temperature (23 ± 1°C), humidity (50 ± 10%), and a 12-h light/dark cycle, with a 1-week acclimation period. A total of 10 mice were randomly divided into 2 groups (5 mice per group) and subcutaneously injected with OC cells (SKOV3/shNC and SKOV3/sh1, 5 × 10⁶ cells per mouse). Separately, 12 mice were randomly divided into 2 groups (6 mice per group) and injected with OC cells (A2780/DHCR24 and A2780/Vector, 5 × 10⁶ cells per mouse). Mice were maintained for 30–40 days (15 July to 25 August 2024). When the subcutaneous tumor diameter reached 1.0 cm, the mice were euthanized with pentobarbital sodium, and solid tumors were surgically extracted and weighed. The tumor tissue was fixed with a tissue-fixative solution for immunohistochemistry (IHC).

Animal experiments were approved by the Institutional Animal Care and Use Committee of Chongqing Medical University (approval number: IACUC-CQMU-2024-0705). The study was reported in accordance with ARRIVE guidelines.

The tissue sections were baked at 60°C. Sodium citrate solution was used for antigen repair. Subsequently, the tissue sections were subjected to a blocking solution and incubated with the primary antibody. A diaminobenzidine (20x) chromogenic solution (Cat. No. ZLI-9018, ZSGB-BIO, Beijing, China) was applied. Nuclei were stained with hematoxylin (Cat. No. G1004, Servicebio, Wuhan, China). Information regarding the antibodies that were used is provided in Table S5.

Transcriptome sequencing was conducted by Seqhealth Technology Co., Ltd. (Wuhan, China) using a NovaSeq 6000 sequencer (Illumina, San Diego, CA, USA). Differentially expressed genes between groups were identified using the ‘edgeR’ package (version 3.42.0) in R 4.3.0 (The R Foundation for Statistical Computing, Vienna, Austria), with screening criteria of |logFC| > 1 and adjusted p-value < 0.05. The results were visualized in a volcano plot using the ggplot2 R package (version 3.4.0). Subsequently, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis and Gene Set Enrichment Analysis (GSEA) were conducted using the “ClusterProfiler” R package (version 4.8.0), with a significance threshold of p < 0.05 [33,34].

SKOV3 cells were treated with TGF-β1 (Cat. No. HY-P7118, MCE, Middlesex, NJ, USA), while A2780 cells were treated with the TGF-β1 pathway inhibitor SB431542 (Cat. No. S1067, Selleck Chemicals, Houston, TX, USA). SKOV3 and ES2 cells were treated with the DHCR24 inhibitor SH-42 (Cat. No. HY-143228, MCE, NJ, USA). Functional assays were performed after 48 h, and total cellular proteins were extracted for Western blot at 72 h. Relevant experimental details can be found in the corresponding methods section. SB431542 and SH-42 were dissolved in dimethyl sulfoxide (DMSO) to a concentration of 10 μM, and TGF-β1 was dissolved in PBS (pH 7.2–7.4) containing bovine serum albumin (BSA) to a concentration of 10 ng/mL.

Statistical analysis and graph plotting were performed using SPSS 25.0 and GraphPad Prism 8.0 software, while image processing was conducted with Fiji software. Continuous variables are expressed as mean ± standard deviation (Mean ± SD), with comparisons between two groups analyzed by t-test and multi-group comparisons by one-way ANOVA. Categorical variables were described as frequency (percentage) and compared using the chi-square test. p < 0.05 was considered statistically significant (*p < 0.05, **p < 0.01, ***p < 0.001), with corrections for multiple comparisons applied.

The GEPIA database indicated that DHCR24 in the OC tissues was higher than that in the normal tissues (Fig. 1a). Kaplan-Meier database showed shorter progression-free survival (PFS) in patients with high DHCR24 expression (Fig. 1b). Subsequently, using qPCR and IHC, we confirmed significant upregulation of DHCR24 expression in the OC tissues (p < 0.001) (Fig. 1c,d). To compare the relationship between DHCR24 and clinicopathological parameters of OC, we used IHC to detect the expression of DHCR24 in 70 cases of OC. Additionally, we examined the expression of TGF-β1 in OC tissues using IHC. The results showed that TGF-β1 expression was significantly higher in the DHCR24 high-expression group than in the DHCR24 low-expression group, suggesting a potential positive correlation between TGF-β1 expression and DHCR24 expression (Fig. 1e). By analyzing the pathological data of patients, we found that the level of DHCR24 was significantly related to the International Federation of Gynecology and Obstetrics (FIGO) stage (p = 0.015) and ascites condition (p = 0.027) of patients with OC. Patient data are listed in Table S1. The qPCR results suggested increased DHCR24 expression in OC cells (All p < 0.05) (Fig. 1f). Collectively, these findings suggested a significant increase in DHCR24 expression in OC, which is linked to undesirable outcomes for OC.

SKOV3 and CAOV3 cells were used to establish an in vitro model with stable and low expression of DHCR24. A stable model of DHCR24 overexpression was established using ES2 and A2780 cells. Results of qPCR and WB confirmed that DHCR24 expression decreased in the cells of the DHCR24 low-expression group (SKOV3/sh1, sh2, sh3) (Fig. 2a,b) but increased in the cells of the DHCR24 high-expression group (ES2/DHCR24) (All p < 0.001) (Fig. 2c,d). These validations were also performed in the other two cell lines (Fig. S1).

CCK-8 and clone formation experiments confirmed that DHCR24 overexpression increased proliferation and colony formation in ES2 cells (Fig. 2e,f), whereas decreased DHCR24 expression had reverse effects in SKOV3 cells (All p < 0.01) (Fig. 2g,h). The same experiments were performed in the other two cell lines, yielding consistent results (Fig. S1).

Flow cytometry results showed that DHCR24 overexpression reduced apoptosis in OC cells (Fig. 3a), whereas decreased DHCR24 expression had reverse effects (Fig. 3b). In addition, WB suggested increased Bcl2 expression but decreased Bax expression in the high DHCR24 expression group (Fig. 3c), and reverse results were noted in the low DHCR24 expression group (All p < 0.05) (Fig. 3d). When the same experiments were performed in the other two cell lines, the conclusions were consistent (Fig. S2).

Transwell experiment indicated that the migration and invasion of OC cells decreased after DHCR24 knockdown (Fig. 4a) but were enhanced in the DHRC24 overexpression group (Fig. 4b). In addition, the wound healing experiment showed that the wound healing rates of OC cells decreased in the knockdown groups (All p < 0.01) (Fig. 4c). The Transwell assay in CAOV3 and A2780 cells yielded results consistent with the above findings (Fig. S3).

To demonstrate the effect of DHCR24 on EMT, we performed WB and found that the levels of E-cadherin in the DHCR24 low-expression group (SKOV3/sh1, sh2) were higher than that in the control group (SKOV3/shNC and SKOV3), whereas the levels of N-cadherin and Vimentin decreased (Fig. 4d). In ES2 cells, compared with the control group (ES2/Vector and ES2), E-cadherin expression in the DHCR24 overexpression group (ES2/DHCR24) decreased, whereas N-cadherin and Vimentin expression increased (All p < 0.001) (Fig. 4e). When the same experiments were performed in the other two cell lines, the conclusions were consistent (Fig. S3). Taken together, these results suggested that DHCR24 promotes EMT in OC cells.

Mice were randomly categorized into the SKOV3/sh1 (DHCR24 low expression), SKOV3/shNC (negative control), A2780/DHCR24 (DHCR24 overexpression), and A2780/Vector (control) groups. The tumor volume, tumor growth rate, and weight decreased (Fig. 5a–d) in the DHCR24 low-expression group. Therefore, we inferred that DHCR24 knockdown inhibits OC cell proliferation in vivo. IHC indicated that DHCR24, Ki67, Bcl2, and N-cadherin expression levels were higher in the SKOV3/shNC group than in the SKOV3/sh1 group (Fig. 5e–h). Additionally, IHC results further showed higher expression of TGF-β, P-Smad2, and P-Smad3 in the SKOV3/shNC group, while E-cadherin expression exhibited the opposite pattern (Fig. S4). Furthermore, tumor volume, tumor growth rate, and weight increased significantly in the DHCR24 overexpression group, suggesting that DHCR24 overexpression promotes OC cell proliferation in vivo (Fig. S5). The IHC results were consistent with those of the SKOV3 group (All p < 0.01) (Fig. S6).

SKOV3/sh1 and SKOV3/shNC cells were utilized for transcriptome sequencing. Heat maps and volcano plots showed differentially expressed genes between SKOV3/sh1 and SKOV3/shNC cells (Fig. 6a,b). Data analyses suggested a significant correlation between DHCR24 and the TGF-β pathway in OC (Fig. 6c,d). Subsequently, we detected key protein molecules of the TGF-β pathway based on WB results. In SKOV3 cells, TGF-β1 and p-Smads expression levels were lower in the DHCR24 low-expression group (SKOV3/sh1, sh2) than those in the control group (SKOV3/shNC, SKOV3); however, Smad2/3 levels remained unchanged (Fig. 6e). The reverse was also true in A2780 cells (All p < 0.01) (Fig. 6f).

In SKOV3 cells, DHCR24 knockdown reduced their malignant behavior, whereas TGF-β1 treatment reversed this effect (Fig. 7a,c,e). In A2780 cells, DHCR24 overexpression enhanced their malignant behavior, but SB431542 treatment reversed this effect (Fig. 7b,d). In addition, WB results indicated that DHCR24 knockdown decreased the levels of key proteins of the TGF-β1 pathway in SKOV3 cells, whereas TGF-β1 treatment reversed this effect (Fig. 7f). In A2780 cells, DHCR24 overexpression increased the levels of key proteins of the TGF-β1 pathway, whereas SB431542 treatment reversed this effect (all p < 0.05) (Fig. 7g). Collectively, these findings indicated that DHCR24 regulates OC cell malignant behavior and correlates with the TGF-β1 pathway.

To investigate whether DHCR24 can serve as a potential therapeutic target for OC, SKOV3, and ES2 cells were treated with the DHCR24 inhibitor SH-42. Results showed that SH-42 reduced the proliferation, migration, and invasion capabilities of SKOV3 (Fig. 8a,b) and ES2 cells (Fig. 8c,d). Further, SH-42 also reversed the malignancy-promoting effects of DHCR24 overexpression in ES2 cells (Fig. 8c,d). In summary, SH-42 can significantly reduce the cancer-promoting effects of DHCR24 in OC cells.