Transcriptome screening and functional validation of FN1 in the regulation of granulosa cell proliferation in polycystic ovary syndrome.

Polycystic ovary syndrome (PCOS) is a common reproductive, endocrine, and metabolic disorder that affects women of reproductive age. It impacts approximately 11%–13% of women worldwide, placing a significant burden on healthcare systems and the economy [ 1 ]. The main clinical features include hyperandrogenemia, polycystic ovarian changes, and ovulatory dysfunction, which can lead to infertility. Other symptoms include metabolic issues such as insulin resistance, obesity, and type 2 diabetes (T2D). Although PCOS is a significant threat to women's reproductive health, its underlying mechanisms are not fully understood, and research in this area remains limited [ 2 ]. Currently, PCOS is diagnosed based on the 2003 Rotterdam Consensus Criteria. Adult patients must meet two of the following three criteria: hyperandrogenemia (clinical or biochemical signs), ovulatory dysfunction (oligomenorrhea or amenorrhea), and polycystic ovarian morphology (confirmed by ultrasound showing ≥ 20 antral follicles per ovary or an anti-Müllerian hormone [AMH] level of ≥ 4.6 ng/ml). For adolescents, hyperandrogenemia combined with menstrual disturbances is required, with ovarian morphological indicators excluded to prevent misdiagnosis [ 3 , 4 ]. Based on these criteria, PCOS can be classified into four distinct phenotypic types. Types A (hyperandrogenemia + ovulatory disorder + polycystic ovaries) and C (hyperandrogenemia + polycystic ovaries) focus on metabolic issues. They are notably linked to increased risks of insulin resistance, T2D, and cardiovascular disease. Types B (hyperandrogenemia + ovulatory disorder) and D (ovulatory disorder + polycystic ovaries without hyperandrogenemia) highlight reproductive problems, which can present as anovulatory infertility, a higher risk of endometrial cancer, and pregnancy complications. Type D often goes unnoticed due to the lack of typical hyperandrogenic signs and requires ongoing monitoring of AMH and ovarian morphology [ 4 ]. The wide variety of clinical symptoms in PCOS makes individualized diagnosis and treatment complex. Metabolic and psychological issues tend to persist throughout the disease, and health risks differ significantly among phenotypes [ 5 ]. Furthermore, current treatments–such as oral contraceptives for regulating menses and metformin for improving insulin sensitivity–only address symptoms and do not target the root causes, like abnormal neuroendocrine regulation and adipose tissue dysfunction. Patients remain at long-term risk for complications, including T2D and endometrial cancer [ 6 ]. Therefore, early detection and management of metabolic issues, along with personalized treatment plans and psychological support, are crucial for improving outcomes. In the pathophysiological process of PCOS, ovarian granulosa cells (GCs) play a vital role [ 7 , 8 ]. These cells play a crucial role in the development of follicles. They not only secrete various growth factors but also transmit key information through cell connections with oocytes, thereby affecting follicle maturation and ovulation [ 9 , 10 ]. Numerous studies have shown that GC dysfunction is associated with disorders of follicular development, including excessive follicular recruitment, impaired selection of dominant follicles, follicular atresia, anovulation, and metabolic issues in PCOS [ 11 - 13 ]. Communication between oocytes and GCs is essential for normal follicular development and steroid hormone secretion by GCs [ 14 ]. Additionally, GCs are considered one of the most effective non-invasive methods for assessing oocyte quality and ovarian function [ 15 ]. In clinical practice, they offer a source of biological samples for studying female follicular development without wasting oocytes, making them a more ethical choice [ 16 ]. However, few studies have employed GC transcriptome sequencing to explore the transcriptome features of GCs in PCOS patients in the Xinjiang region of China. Therefore, investigating the role of GCs in the pathogenesis of PCOS could deepen our understanding of its biological basis and aid in the development of relevant treatment strategies. As molecular biology and genomics technologies have advanced, transcriptomics has become a valuable tool for studying gene expression and is now widely used in biomedical research. Transcriptomics can thoroughly analyze changes in intracellular transcriptional activity using high-throughput sequencing technologies, thereby uncovering gene expression patterns under various physiological and pathological conditions. In recent years, an increasing number of studies have utilized transcriptome sequencing of GCs to investigate the pathogenesis of PCOS [ 17 - 19 ]. However, owing to the complex heterogeneity of clinical phenotypes and ethnic diversity associated with PCOS, differences in GC transcriptome features may exist between PCOS patients and non-PCOS controls in various regions. Thus, screening for differentially expressed genes (DEGs) in GCs from PCOS patients and non-PCOS controls in the Xinjiang region of China—using transcriptome sequencing and considering regional and ethnic factors—can help identify new mRNA expression profiles and discover novel biomarkers. This approach is crucial for targeted classification, diagnosis, and personalized treatment of PCOS, providing valuable insights for both research and clinical practice. In this study, we performed transcriptome sequencing to analyze the gene expression profiles of GCs from patients with PCOS and healthy females in the Xinjiang region of China. Differential gene expression analysis was then conducted on the PCOS and non-PCOS control groups, followed by functional interaction prediction analysis. The results showed that NPTX2, FN1, CCN1, IDH1, ZCCHC17, and CREG1 were closely linked to the development of PCOS. Additionally, it was observed that knocking down FN1 in KGN cells suppressed GC proliferation and caused disruptions in cell cycle distribution. By identifying the DEGs associated with PCOS, we can gain a deeper understanding of the roles these genes play in GC function and their impact on follicular development, hormone production, and overall ovarian health. This understanding will help clarify the molecular mechanisms behind PCOS and support further research into its pathology.

GC source: GCs were collected from 32 infertile female patients undergoing IVF/ICSI treatment at the First Affiliated Hospital of Shihezi University between October 2023 and October 2024. The patients, aged 25 to 40 years, were divided into two groups: a PCOS group of 16 patients and a non-PCOS control group of 16 patients who were infertile due to male or tubal factors. Inclusion criteria: The diagnostic criteria for PCOS were based on the internationally recognized 2003 Rotterdam Consensus Criteria [ 20 ], requiring at least two of the following conditions: (1) Clinical and/or biochemical signs of hyperandrogenemia. Hyperandrogenemia is defined as a fasting testosterone level on the third day of menstruation exceeding the normal range. Clinical features may include acne and hirsutism (excluding other diseases that can cause hyperandrogenemia, such as congenital adrenal hyperplasia, androgen-secreting tumors, and Cushing's syndrome); (2) oligo-ovulation and/or anovulation, with oligo-ovulation defined as a menstrual cycle longer than 35 days; (3) polycystic ovarian morphology: one or both ovaries having ≥ 2 antral follicles measuring 2–9 mm in diameter or an ovarian volume > 10 ml (applied per institutional clinical protocol at the time of recruitment; note that the updated 2023 ESHRE guideline recommends a threshold of ≥ 20 antral follicles per ovary). Non-PCOS control group: Infertile patients undergoing IVF/ICSI during the same period, with infertility caused by tubal or male factors. Exclusion criteria: Oligomenorrhea or hyperandrogenemia caused by other factors; diseases affecting oocytes, such as endometriosis, decreased ovarian reserve, chromosomal translocation, recurrent miscarriage, and fragile X syndrome. This study was approved and conducted by the Ethics Committee of the Department of Reproductive Medicine at the First Affiliated Hospital of Shihezi University (Ethics approval number: KJ2022-178-01). Written informed consent was obtained from all patients before the collection of samples. All patients were treated with the standard long protocol of a GnRH agonist for ovulation induction. Gonadotropin (Gn) was discontinued when at least three dominant follicles measuring ≥ 16 mm in diameter were present in both ovaries, or when two follicles measuring ≥ 17 mm or one follicle measuring ≥ 18 mm in diameter were present. Human chorionic gonadotropin was injected intramuscularly at 24:00 on the same day. After 36 h, oocytes were retrieved under transvaginal ultrasound guidance. Follicular fluid was aspirated through the posterior fornix using a 16-gauge double-lumen oocyte retrieval needle. The oocytes were carefully collected from each tube of aspirated follicular fluid and quickly transferred to a G-MOPS medium (Vitrolife). The remaining follicular fluid was collected in a sterile centrifuge tube and centrifuged at 500 g for 10 min. The recovered GCs were washed with 5 ml of phosphate-buffered saline (1× PBS) and centrifuged at 1,000 g for 10 min. The GCs were then isolated and purified using a Percoll (Pharmacia) density gradient centrifugation process. The cells were dissolved in 1 ml of Trizol (Solarbio) and immediately stored at –80°C for further analysis. We randomly selected five pairs (5 PCOS and 5 non-PCOS control) from the samples collected for transcriptome sequencing. Total RNA was isolated and purified from the samples using TRIzol (ThermoFisher, 15596018), following the manufacturer's protocol. The quantity and purity of the total RNA were then assessed using a NanoDrop ND-1000 (NanoDrop Technologies), and the integrity of the RNA was verified with a Bioanalyzer 2100 (Agilent Technologies). Samples needed concentrations of > 50 ng/μl, RNA integrity number values of > 7.0, and a total RNA concentration of > 1 μg for downstream experiments. All sequencing samples were processed in the same batch to minimize potential batch effects. Two rounds of purification were performed using oligo(dT) magnetic beads (Dynabeads Oligo(dT), cat. 25-61005, Thermo Fisher Scientific). The captured mRNA was fragmented with a magnesium ion fragmentation kit (NEBNext Magnesium RNA Fragmentation Module, catalog number E6150S) at 94°C for 5–7 min. Then, cDNA was synthesized from the fragmented RNA using reverse transcriptase (Invitrogen SuperScript II Reverse Transcriptase, catalog number 18966-49). Subsequently, E. coli DNA polymerase I (NEB, catalog number m0209) and RNase H (NEB, catalog number m0297) carried out two-strand synthesis, converting the RNA and DNA complex into double-stranded DNA. During this process, dUTP was incorporated into the DNA strands (Thermo Fisher, catalog number R0133). The double-stranded DNA ends are then complemented to blunt ends, and an A base is added to each end to facilitate ligation with a linker containing a T base. The fragment size is subsequently screened and purified using magnetic beads. The double strand is digested with UDG enzymes (NEB, cat. M0280), followed by PCR pre-denaturation at 95°C for three minutes, then eight cycles of denaturation at 98°C for 15 sec each, annealing at 60°C for 15 sec, extension at 72°C for 30 sec, and a final extension at 72°C for five minutes, resulting in a library with fragment sizes of approximately 300 bp ± 50 bp (strand-specific library). Finally, Illumina Novaseq 6000 (LC Biotechnology Co., Ltd.) was used according to standard procedures in PE150 sequencing mode. The cDNA library, prepared from RNA of a pool of granule cells from the PCOS and non-PCOS control groups, was sequenced on the Illumina Novaseq 6000 platform with a read length of 150 bp. The raw sequencing data contained linker sequences and low-quality bases, which could impact subsequent assembly and analysis. Therefore, Cutadapt ( https://cutadapt.readthedocs.io/en/stable/; version 1.9) was used to clean the data. The settings included: 1) removing reads with linker sequences; 2) removing reads with polyadenylate (polyA) and polyguanylate (polyG) sequences; 3) removing reads with more than 5% unknown bases (N); and 4) removing reads with more than 20% low-quality bases (Q ≤ 20). The quality of the cleaned data was verified using FastQC ( http://www.bioinformatics.babraham.ac.uk/projects/fastqc/ , v0.11.9), focusing on Q20, Q30, and GC content metrics. A high-quality, paired-end, cleaned read data totaling 63.85 Gbp was obtained, and a total of 40,956 genes were identified. The original sequencing data have been uploaded to the NCBI Gene Expression Omnibus under the accession number GSE304677. Sample correlation analysis: R software was used to perform a correlation analysis of the samples. Correlation analysis of two parallel experiments helps assess the reliability of the results and the stability of the process. Calculating the Pearson correlation coefficient between two replicate measures of the reproducibility between samples. The closer the correlation coefficient is to 1, the better the reproducibility between the two experiments. Analysis of DEGs: Gene differential expression analysis was performed between the two groups using the R package DESeq2. Refer to the supplementary materials for the gene analysis results. Genes with a false discovery rate (FDR) less than 0.05 and an absolute fold change of at least 2 were considered DEGs. DEGs are defined as genes with |log₂(fold change)| ≥ 1 and q < 0.05. Enrichment analysis: To clarify the biological functions of the DEGs, we performed Gene Ontology (GO) function and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses on the DEGs. GO terms and pathways with a p-value less than 0.05 were considered significantly enriched among the DEGs. These were annotated and visualized using the clusterProfiler and ggplot2 packages, respectively. Additionally, Gene Set Enrichment Analysis (GSEA) was conducted on the DEGs using the clusterProfiler package in R software (version 4.2.1). Enriched pathways identified by GSEA were selected based on the following criteria: |Normalized Enrichment Score (NES)| > 1, FDR (q-value) < 0.25, and p < 0.05. Construction of protein-protein interaction (PPI) network and identification of hub genes: The online analysis tool STRING (Search Tool for the Retrieval of Interacting Genes) database was used to analyze PPI, setting a confidence score above 0.40 to screen PPI pairs. Based on this, the PPI network was built using Cytoscape software (version 3.9.1) to illustrate its topological structure. The top 10 hub genes were identified using the BottleNeck algorithm within the Cytohubba plugin of Cytoscape software (3.9.1). KGN cells were obtained from the National Collection of Authenticated Cell Cultures of China and cultured in Dulbecco’s Modified Eagle Medium/F-12 medium supplemented with 10% fetal bovine serum, 100 U/ml penicillin, and 100 μg/ml streptomycin. The cells were incubated at 37°C with 5% CO2. They were passaged every 2–3 days, and cells at around the third passage were used for experiments. To generate a stable FN1 knockdown in KGN cells, four shRNA sequences (shFN1-1 to shFN1-4; detailed sequences are listed in Supplementary Table 1 ) targeting the FN1 gene were designed. These sequences were cloned into the pLKO.1-CMV-copGFP-PURO lentiviral vector by Tsingke to create recombinant interference plasmids. The four shFN1 plasmids were then transfected into 293T cells using Lipofectamine 3000 (Thermo Fisher Scientific). Cells were harvested 48 h post-transfection, and shFN1-3, which showed the highest knockdown efficiency, was selected for further experiments via RT-qPCR. The chosen shFN1 plasmid was co-transfected into 293T cells with the lentiviral packaging plasmid pMD2.G and helper plasmid psPAX2 at a 2:1:2 ratio. Viral supernatants were collected twice, at 72 and 96 h after transfection. KGN cells were seeded in 6-well plates at a density of 2 × 10 5 cells per well. When cell confluence reached 30%–40%, the viral supernatant was mixed 1:1 with target cell culture medium and added to the cells. The medium was replaced 12 h after infection, and the cells were continuously selected with puromycin for seven days. The knockdown efficiency of FN1 was verified through copGFP fluorescence and RT-qPCR. Finally, protein-level silencing was confirmed by Western blot analysis, with data compared to the shRNA negative control group (sh-NC). Total cellular RNA was extracted and purified using TRIzol reagent (Solarbio) according to the manufacturer's instructions. The primers ( Table 1 ) were designed and synthesized by Beijing Tsingke Biotechnology Co., Ltd. Five pairs of GC samples were subjected to RT-qPCR validation using the miRCURY LNA Universal RT MicroRNA PCR System (Exiqon), according to the manufacturer's instructions. The results were analyzed using an ABI 7900 HT (Applied Biosystems). The PCR reaction program consisted of pre-denaturation at 95°C for 1 min, denaturation at 95°C for 10 sec, and annealing at 60°C for 20 sec, repeated 40 times. The expression levels of each gene were normalized to glyceraldehyde 3-phosphate dehydrogenase (GAPDH) and analyzed using the 2-ΔΔCt method. Each sample in each group was measured in triplicate, and the experiment was repeated three times. The treated KGN cells were washed with cold PBS and then lysed on ice for 30 min using RIPA lysis buffer (Beyotime, P1005), which contained protease and phosphatase inhibitors (Beyotime, P0013). The supernatant was collected by centrifugation at 12,000 g for 15 min at 4°C. Protein concentration was measured using the BCA method (Beyotime, P0010) and adjusted to 2 μg/μl. Thirty micrograms of protein were separated by 10% SDS-PAGE and transferred to a PVDF membrane (Millipore, IPVH00010) via wet transfer. The PVDF membrane was blocked by soaking in TBST containing 5% skimmed milk powder at room temperature for one hour and incubated overnight at 4°C with the following primary antibodies: anti-FN1 (1:1,000, Santa Cruz, 15613-1-AP), anti-phospho-Akt (Ser473) (1:1,000, Cell Signaling Technology, #4060), anti-total Akt (1:1,000, Cell Signaling Technology, #4691), anti-cyclin D1 (1:1,000, Cell Signaling Technology, #2978), anti-p21 (1:1,000, Cell Signaling Technology, #2947), anti-p27 (1:1,000, Cell Signaling Technology, #3686), and anti-GAPDH (1:5,000, Proteintech, 10494-1-AP) as a loading control. After washing three times with TBST, the membrane was incubated with HRP-conjugated secondary antibody (Servicebio, GB21301) on a shaker at 37°C for two hours. The enhancer solution and the stable peroxidase solution in the ECL reagent were mixed at a 1:1 ratio. The working solution was applied to the PVDF membrane, which was then placed in a chemiluminescence imager for development and imaging. The grayscale value of the target band was quantified using Image Lab software and normalized to GAPDH as an internal control. All experiments were performed with three biological replicates, and a sh-NC control group was included to ensure the reliability of the results. Lentivirus was used to stably infect cells in the logarithmic growth phase and a healthy growth state, with the cell density adjusted to 1 × 10 5 cells/ml. The cells were seeded into a 96-well plate at a volume of 100 μl per well. After 48 h, 10 μl of enhanced CCK8 reagent (Beyotime, C0041) was added to each well. The plate was then incubated at 37°C for one hour, and the absorbance (OD 450) of each well was measured using a microplate reader. The Click-iT EdU-555 Cell Proliferation Assay Kit (Servicebio, G1602) was used to assess the proliferation activity of KGN cells. The specific steps were as follows: Lentivirus-stably transfected KGN cells were seeded into 48-well plates at a density of 1 × 10 5 cells/ml (300 μl/well). After culturing at 37°C for 48 h until the confluence reached 70%–80%, 50% of the original medium was aspirated, and an equal volume of pre-warmed 50 μM EdU working solution (diluted from a 10 mM EdU stock solution with complete medium) was added, then incubated for an additional 2 h. The EdU-containing medium was discarded, and the cells were washed three times with pre-cooled PBS (each wash lasting five minutes). Next, 4% paraformaldehyde (containing 0.1 M PBS, pH 7.4) was added for fixation at room temperature for 20 min. After washing with PBS three times, 2 mg/ml glycine was used for neutralization for 10 min. Then, 0.5% Triton X-100 (Beyotime, ST797) was added for permeabilization at room temperature for 20 min, followed by three PBS washes. The Click-iT reaction mixture was prepared following the kit instructions, with 200 μl added to each well and incubated in the dark for 30 min. The cells were then washed three times (five minutes each) sequentially with PBS containing 0.5% Triton X-100, pre-cooled methanol, and PBS. Hoechst 33342 was diluted 1:1,000 in PBS, and the cells were stained for 15 min before being mounted. Images were captured using a fluorescence microscope (Olympus CKX53). The percentage of EdU-positive cells in three fields of view per well was quantified using ImageJ software (National Institutes of Health). Three replicate wells were prepared for each group, and the experiments were repeated three times. The propidium iodide (PI) staining system in the Annexin V-APC/PI Apoptosis Kit (Elabscience, E-CK-A217) was used to analyze cell cycle distribution. Lentivirus-transfected KGN cells were seeded in a 6-well plate at a density of 1 × 10⁵ cells/ml (2 ml per well). After 48 h of transfection, the cells were digested with trypsin, collected, and then centrifuged at 1,000 rpm for 5 min. They were washed twice with pre-cooled PBS. After resuspending the cells in 300 μl of PBS, 1.2 ml of pre-cooled absolute ethanol was slowly added, and the cells were fixed at 4°C for 2 h. Centrifugation was used to remove the ethanol, after which the cells were resuspended in 1 ml of PBS and equilibrated at room temperature for 15 min. The cells were then centrifuged at 1,000 rpm to discard the supernatant, and 100 μl of RNase A (100 μg/ml, prepared in PBS) was added to digest RNA at 37°C for 30 min. Next, 400 μl of PI staining solution was added, and the cells were incubated in the dark for 30 min. Finally, the cell suspension was filtered through a 40 μm filter membrane, and cell cycle analysis was performed using a flow cytometer (Beckman Coulter, CytoFLEX). Apoptotic cell death was quantified using the Cell Death Detection ELISA kit (Roche, 11544675001), which measures cytoplasmic histone-associated DNA fragments (mono- and oligonucleosomes). KGN cells transfected with shFN1 or sh-NC lentivirus were seeded in 96-well plates at a density of 1 × 10⁴ cells per well and cultured for 48 h. Following the manufacturer's protocol, cells were lysed with 200 μl lysis buffer and incubated at room temperature for 30 min. The lysate was centrifuged at 200 g for 10 min, and 20 μl of the supernatant from each sample was transferred to a streptavidin-coated microplate. Subsequently, 80 μl of immunoreagent containing anti-histone-biotin and anti-DNA-peroxidase antibodies was added to each well. After incubation at room temperature for 2 h with gentle shaking, wells were washed three times with incubation buffer. The peroxidase substrate ABTS was added, and the plate was incubated for 10–20 min until color development was sufficient. Absorbance was measured at 405 nm (reference wavelength 490 nm) using a microplate reader. The enrichment factor of nucleosomes in the cytoplasm was calculated relative to the control group. Data analysis and visualization were performed using SPSS 27.0 (IBM) and GraphPad Prism 9 (GraphPad Software). The normality of continuous variables was evaluated using the Kolmogorov–Smirnov test. Data conforming to a normal distribution were presented as the mean ± standard deviation (SD), and two-sided Student's t-tests were used for intergroup comparisons (homogeneity of variance was confirmed by the Levene test). The Mann–Whitney U-test was used to analyze non-normally distributed data, which were presented as median (interquartile range). Categorical variables were described by frequency (n) and percentage (%), and intergroup differences were analyzed using the chi-square test or Fisher's exact test (when the theoretical frequency was less than 5). All experimental data were based on at least three independent biological replications. The threshold for statistical significance was set at two-sided p < 0.05, and the error bars in the figures represent the SD or standard error of the mean (SEM).

The overall study design and analytical workflow are illustrated in Supplementary Fig. 1 . The baseline characteristics of the patients are listed in Table 2 . There were no significant differences in mean age, body mass index, infertility type, years of infertility, basal follicle-stimulating hormone (FSH) level, basal E2 level, total cholesterol, triglycerides, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, fasting blood glucose, fasting insulin, homeostatic model assessment of insulin resistance, or fertilization method between the two groups (p > 0.05). AMH, basal luteinizing hormone (LH) level, LH/FSH ratio, and the number of retrieved oocytes in the PCOS group were significantly higher than in the non-PCOS control group. Conversely, the percentage of good embryos on D3 in the PCOS group was notably lower than in the non-PCOS control group (p < 0.05). To identify DEGs, transcriptome sequencing was performed on GC samples from five PCOS patients and five non-PCOS control patients. As shown in Fig. 1A , Pearson's correlation coefficient plots were used to assess relationships between samples and confirm the consistency of the biological replicates. Fig. 1B presents a volcano plot illustrating the differential expression of mRNAs between the PCOS and non-PCOS control groups. As shown in Fig. 1C , a bar graph indicates that 157 mRNAs are up-regulated, while 71 mRNAs are down-regulated in PCOS. Fig. 1D displays the hierarchical clustering heat map of differentially expressed RNAs. To analyze the biological classification of DEGs, functional and pathway enrichment analyses were conducted on 228 DEGs using the Database for Annotation, Visualization, and Integrated Discovery (DAVID) tool to identify their potential functions. GO enrichment analysis was performed across three categories: biological processes (BP), cellular components (CC), and molecular functions (MF). The results are presented as a scatter plot in Fig. 2A and as a detailed bar plot in Supplementary Fig. 2A . As shown in Supplementary Fig. 2A , GO analysis revealed that BPs in DEGs were most significantly enriched in signal transduction, cell adhesion, apoptotic processes, cell differentiation, lipid metabolic processes, negative regulation of cell population proliferation, positive regulation of transduction by RNA polymerase II, positive regulation of cell proliferation, positive regulation of cell migration, and positive regulation of gene expression. CCs are primarily enriched in the membrane, cytoplasm, plasma membrane, extracellular region, cytosol, nucleus, extracellular exosomes, extracellular space, nucleoplasm, and collagen-containing extracellular matrices. MFs of DEGs were mainly enriched in protein binding, metal ion binding, identical protein binding, hydrolase activity, transferase activity, nucleotide binding, ATP binding, signaling receptor binding, calcium ion binding, and zinc ion binding. Among all GO terms, protein binding (GO:0005515) exhibited the highest gene count and the lowest enrichment p-value, as illustrated in the scatter plot ( Fig. 2A ). KEGG pathway enrichment analysis classified DEG-associated pathways into six functional groups, with the bar plot presented in Supplementary Fig. 2B . Within the cellular processes, environmental information processing, genetic information processing, human diseases, metabolism, and organic systems. The primary pathways involved in ecological information processing enrichment included focal adhesion, regulation of the actin cytoskeleton, the p53 signaling pathway, and the peroxisome proliferator-activated receptor pathway. The key environmental information processing pathways included the PI3K-Akt signaling pathway, the cAMP signaling pathway, the MAPK signaling pathway, the cytokine-cytokine receptor interaction pathway, and the TGF-beta signaling pathway. Pathways enriched in genetic information processing were the proteasome, RNA degradation, mRNA surveillance pathway, and ubiquitin-mediated proteolysis. The top pathways in the 'Proteoglycans in cancer' category were MicroRNAs in cancer, Pathways in cancer, Amoebiasis, and AGE-RAGE signaling pathways in diabetic complications. Metabolic pathways identified included glyoxylate and dicarboxylate metabolism, carbon metabolism, propionate metabolism, and nicotinate and nicotinamide metabolism. The most critical systemic pathways for organismal enrichment include axon guidance, salivary secretion, regulation of TRP channels by inflammatory mediators, complement and coagulation cascades, and aldosterone synthesis and secretion. The KEGG scatter plot ( Fig. 2B ) further highlights that the PI3K-Akt signaling pathway, focal adhesion, cAMP signaling pathway, ErbB signaling pathway, TGF-beta signaling pathway, NF-kappa B signaling pathway, JAK-STAT signaling pathway, steroid hormone biosynthesis, epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor resistance, and insulin resistance represent the most significantly enriched pathways by rich factor and gene count. GSEA was performed to further validate pathway-level enrichment at the gene set level. A total of six gene sets met the predefined significance thresholds. As illustrated in Fig. 2C–E , the TGF-beta signaling pathway, the MAPK signaling pathway, and the PI3K-Akt signaling pathway were significantly enriched in the PCOS group relative to controls. An additional three enriched gene sets are presented in Supplementary Fig. 3 : focal adhesion, the Hippo signaling pathway, and the p53 signaling pathway ( Supplementary Fig. 3C ). All six pathways demonstrated positive enrichment scores, indicating coordinated upregulation of the corresponding gene sets in PCOS GCs compared to non-PCOS infertile controls. The PPI networks were constructed using the online database STRING. The full PPI network, comprising all 228 DEGs, is presented in Supplementary Fig. 4 , in which orange nodes represent upregulated genes and green nodes represent downregulated genes. The BottleNeck algorithm within the CytoHubba plug-in was applied to rank nodes by topological importance, and the top 25 hub genes are illustrated in Fig. 3A . Node color intensity reflects the relative hub score, with deep red indicating the highest centrality. Among the identified hub genes, FN1 occupied the most central position in the network, exhibiting the greatest number of interactions with other hub candidates including CCN1, IL1B, THBS1, SERPINE1, AREG, RUNX2, CAV1, CXCR4, and VCL, confirming FN1 as the predominant hub gene within the PCOS GC DEG network. Further analysis based on KEGG pathway enrichment revealed that FN1 was significantly enriched in the PI3K-Akt signaling pathway. As illustrated in Supplementary Fig. 5 , FN1, annotated as a component of the extracellular matrix (ECM, highlighted in red), interfaces with the PI3K-Akt pathway through integrin heterodimers (ITGA/ITGB) and focal adhesion kinase (FAK), thereby converging on downstream AKT phosphorylation and its effector cascades governing cell proliferation, survival, and cell cycle progression. To verify the interactions among the genes that were significantly upregulated in PCOS GCs, protein interaction analysis was performed using the STRING online database on genes such as FN1, ADAMTS1, THBS1, CCN1, SEMA3A, ITGA2, HTRA1, FDX1, NPTX2, AREG, and others. Fig. 3B shows that FN1 is closely related to ADAMTS1, THBS1, AREG, and other genes. These results suggest that FN1 may play a key regulatory role in oocyte maturation by influencing these BP. Nine genes that were differentially expressed were selected for validation based on their significance in the pathophysiology of PCOS. These genes were NPTX2, GSTA1, FN1, CCN1, SERPINE1, IDH1, ZCCHC17, CREG1, and DBI ( Fig. 4 ). Compared to the non-PCOS control group, the expression of NPTX2, FN1, and CCN1 mRNA in the GCs of PCOS patients was significantly increased (**p < 0.01), while the expression of IDH1, ZCCHC17, and CREG1 mRNA was notably decreased (**p < 0.01; **p < 0.001). To investigate the role of FN1 in GCs, we used lentivirus to knock down FN1 in KGN cells. After infecting the KGN cell line with the lentivirus, both mRNA and protein levels of FN1 in KGN cells were significantly reduced, as shown by qRT-PCR and Western blot ( Fig. 5A, B ). CCK-8 experiments demonstrated that FN1 knockdown significantly inhibited the proliferation of KGN cells ( Fig. 5C ). EdU staining revealed a marked decrease in the proportion of EdU-positive cells after knockdown, indicating reduced cell proliferation activity ( Fig. 5D ). Flow cytometry analysis further showed that the proportion of G1-phase cells increased significantly in the knockdown group. In contrast, the proportion of S-phase cells remained unchanged, while the proportion of G2-phase cells significantly decreased ( Fig. 5E ). The cell death detection ELISA assay demonstrated that FN1 silencing significantly increased apoptotic cell death in GCs ( Fig. 6A ). The enrichment factor of cytoplasmic nucleosomes, which reflects the degree of DNA fragmentation during apoptosis, was markedly elevated in the shFN1 group compared to the sh-NC control group. This substantial increase in nucleosome release indicates enhanced apoptotic activity following FN1 depletion. These results demonstrate that FN1 plays a protective role in maintaining GC viability and preventing programmed cell death under standard culture conditions. The ELISA-based quantification provides robust evidence that loss of FN1 expression compromises cell survival mechanisms in GCs. Western blot analysis revealed that FN1 knockdown significantly affected multiple signaling cascades involved in cell proliferation and survival ( Fig. 6B ). Phosphorylated Akt levels at Ser473 were substantially reduced in FN1-depleted cells compared to controls, while total Akt expression remained unchanged. This finding is consistent with a potential association between FN1 expression and PI3K-Akt signaling activity, though whether FN1 acts upstream of this pathway as a direct regulator requires further mechanistic validation, including rescue and pharmacological modulation experiments. Analysis of cell cycle regulatory proteins demonstrated substantial alterations consistent with the observed cell cycle arrest. Cyclin D1 expression was markedly decreased in the shFN1 group, supporting the G1 phase accumulation observed in flow cytometry analysis. This reduction in cyclin D1 levels provides molecular evidence for the impaired G1 to S phase transition following FN1 depletion. Conversely, the cyclin-dependent kinase inhibitor p21 showed significant upregulation, while p27 levels were substantially elevated in FN1-knockdown cells. The coordinated downregulation of pro-proliferative signals and upregulation of cell cycle inhibitors provides mechanistic insight into how FN1 depletion leads to growth arrest in GCs. These molecular changes are consistent with a model in which FN1 expression is functionally associated with both proliferative and survival signaling in GCs, though the precise regulatory mechanisms remain to be formally established.

In the present study, transcriptome sequencing of GCs identified 228 DEGs between PCOS patients and non-PCOS infertile controls, with enrichment in the PI3K-Akt, MAPK, and TGF-β signaling pathways. Among these, FN1 emerged as a central hub gene and was significantly upregulated in PCOS GCs, consistent with prior bioinformatics analyses in comparable cohorts. In our study, we found that several genes were differentially expressed in the GCs of patients with PCOS compared to those of non-PCOS control s. FN1, a key gene, was significantly upregulated in GCs of PCOS patients. High expression of FN1 in gastric cancer mainly occurs in the stroma and is significantly linked to advanced age, high pathological grade, vascular/lymphatic invasion, and an advanced TNM stage. High FN1 expression in the epithelium is associated with a shorter overall survival, while FN1 in the stroma (S-FN1) is linked to tumor progression [ 21 ]. Furthermore, FN1 is notably overexpressed in pancreatic ductal adenocarcinoma and encourages tumor progression by activating the PI3K-AKT signaling pathway. FN1, COL10A1, and FAP are common marker genes for both pancreatic cancer and SARS-CoV-2 infection, and their high expression levels are linked to alterations in the tumor immune microenvironment and variations in drug sensitivity [ 22 , 23 ]. Additionally, FN1 was significantly overexpressed in the serum and lung tissues of chronic obstructive pulmonary disease patients, showing a positive correlation with lung function parameters (FEV1/FVC). Animal models have shown that FN1 promotes the proliferation of bronchial epithelial cells and inhibits apoptosis by regulating AKT phosphorylation levels, indicating that it plays a crucial role in lung tissue remodeling [ 24 ]. Recent studies have indicated that FN1, a key component of the ECM, also influences the reproductive system. FN1 in porcine follicular fluid has been shown to promote nuclear and cytoplasmic maturation of oocytes by activating the PI3K pathway. This suggests that mechanical stress may influence follicular development through FN1-mediated transmission of mechanical signals [ 25 ]. FN1 expression was significantly increased in the ovaries of Nandan Yao chickens with high egg production and was positively associated with serum estrogen levels [ 26 ]. Additionally, FN1 levels are notably higher in the plasma of pre-eclamptic patients and correlate positively with the severity of the disease. Further in vitro studies confirm that FN1 could be a potential biomarker and therapeutic target for vascular endothelial injury [ 27 ]. Several studies have also reported a correlation between abnormal FN1 expression and PCOS. For example, some research indicates that FN1 expression is significantly elevated in the ovaries of PCOS rats induced by DHEA (dehydroepiandrosterone), leading to collagen buildup and fibrosis. This implies that FN1 might be activated during the fibrotic process. However, metformin and simvastatin can indirectly decrease FN1 expression by inhibiting the TGF-β pathway, thereby reducing fibrosis [ 28 ]. In this study, we found that FN1 was significantly overexpressed in PCOS GCs, which aligns with bioinformatics analysis results from the studies by Zhang et al . [ 28 ] and Zhou et al . [ 29 ]. Notably, a clinical intervention study [ 30 ] showed that FN1 could be a potential biomarker for PCOS treatment response when used with GLP-1RA, metformin, and CPA/EE, as well as through plasma proteomic analysis. GLP-1 receptor agonists can lower serum FN1 levels and enhance insulin resistance in patients with PCOS, aligning with the upregulation of FN1 in GCs observed in this study. These results suggest that PCOS may be a key mediator of metabolic disorders and may play a role in changes to the cell microenvironment associated with PCOS. However, our results differ from those reported by Patil et al . [ 31 ], who observed downregulation of FN1 in granulosa luteal cells (GLCs). This difference may stem from the functional stage heterogeneity of the samples: their study focused on luteal GLCs, while this study examined GC samples (GCs). These two cell types are at different stages of development in the ovary and have distinct biological roles in PCOS. High FN1 expression in GCs may contribute to abnormal follicular stromal fibrosis, while FN1 downregulation in GLCs is linked to luteal dysfunction phenotypes. These findings suggest that FN1 shows stage-specific regulatory patterns in PCOS, a hypothesis that could be tested by analyzing FN1 dynamics in GCs and GLCs from the same patient. Additionally, protein interaction network analysis of DEGs significantly upregulated in PCOS GCs revealed that FN1 is closely associated with ADAMTS1, THBS1, AREG, and other genes. ADAMTS1 is an ECM metalloproteinase that plays a vital role in ovulation. Studies have shown that its expression level in the GCs of PCOS patients is linked to GC function, oocyte quality, and embryo development [ 32 ]. Platelet receptor hormone 1 (THBS1) has also been reported to play a key role in promoting follicular atresia and GC apoptosis, which impacts successful ovulation and luteal formation [ 33 , 34 ]. AREG is a member of the EGF family. It is secreted by GCs and theca cells and activates EGFR mainly through an autocrine/paracrine mechanism. AREG regulates follicular development, ovulation, and luteal function. Studies have found that the level of AREG in the follicular fluid of PCOS patients is positively correlated with embryo quality, suggesting that it can be used as a biomarker to predict the success rate of in vitro maturation [ 35 ]. Taken together, the results suggest that FN1 may play a pivotal regulatory role in oocyte maturation by interacting with these genes. The cause of PCOS remains unknown. However, many studies have demonstrated that abnormal proliferation of GCs is strongly linked to the development of PCOS and is also the primary reason for anovulation and infertility in women with PCOS [ 36 - 38 ]. To better understand the role of FN1, we investigated its effect on GC proliferation by knocking down FN1 in KGN cells, a human GC line. CCK-8 and EdU assays demonstrated that FN1 knockdown significantly suppressed KGN cell growth and reduced the EdU-positive ratio, indicating a decrease in proliferative capacity in GCs. Given the increased FN1 expression in PCOS, this functional evidence suggests that FN1 promotes abnormal proliferation, which may contribute to PCOS development by disrupting the cell cycle machinery. This finding aligns with earlier research indicating that FN1 influences ovarian function and follicle development by regulating cell adhesion and migration [ 39 ]. Flow cytometry revealed that the proportion of cells in the G1 phase increased, while the proportion in the G2 phase decreased after FN1 silencing. This supports the mechanism by which FN1 influences the cell cycle by regulating the transition from the G1 phase to the S phase [ 23 ]. It also confirms that FN1 plays a vital role in maintaining the dynamic balance of GCs concerning cell cycle distribution, which is essential for normal follicle development. Our apoptosis analysis revealed that FN1 depletion increased apoptotic cell death in GCs, as evidenced by elevated cytoplasmic nucleosome levels. This observation suggests that FN1 contributes to cell survival in addition to its effects on proliferation. The concurrent decrease in phosphorylated Akt (Ser473) following FN1 knockdown is consistent with reduced PI3K-Akt signaling activity, suggesting a potential functional association between FN1 expression and this survival pathway, though a direct regulatory relationship has not been formally established in the present study. The PI3K-Akt pathway regulates apoptosis through phosphorylation-dependent mechanisms, and its disruption has been associated with increased programmed cell death in various cell types. The reduction in Akt phosphorylation observed in our experiments is consistent with diminished PI3K-Akt signaling activity, which may contribute to the increased apoptosis following FN1 silencing. Analysis of cell cycle regulatory proteins revealed decreased cyclin D1 expression and increased p21 and p27 levels in FN1-depleted cells. These changes are consistent with the G1 phase arrest observed by flow cytometry. Cyclin D1 is required for G1 to S phase transition, while p21 and p27 inhibit cyclin-dependent kinases that drive cell cycle progression. The altered expression of these proteins following FN1 knockdown indicates that FN1 influences multiple components of the cell cycle machinery. These molecular changes may explain how FN1 contributes to the proliferative abnormalities observed in PCOS GCs. The significantly lower Day-3 high-quality embryo rate observed in the PCOS cohort (0.45 vs . 0.69, p < 0.05; Table 2 ) may, at least in part, reflect the molecular alterations in GC function documented in the present study. GCs are indispensable regulators of oocyte developmental competence through direct gap junction–mediated communication and paracrine factor secretion. Elevated FN1 expression in PCOS GCs could compromise the integrity of this communication network by promoting aberrant integrin-mediated signaling and altering the ECM composition of the follicular microenvironment. While a direct causal relationship between FN1 overexpression and reduced embryo quality cannot be established from the current data, this association warrants investigation in future studies correlating individual patient FN1 expression levels with oocyte maturation rates and embryo developmental outcomes. These findings contribute to understanding the molecular mechanisms underlying PCOS pathogenesis. The observed relationship between FN1 expression and GC proliferation suggests that FN1 may represent a candidate target for therapeutic intervention in PCOS. However, the potential clinical utility of targeting FN1 requires validation through in vivo studies and clinical trials. The mechanisms by which FN1 modulation might affect ovarian function in PCOS patients remain to be fully elucidated. Further investigation is needed to determine whether pharmacological manipulation of FN1 expression or activity could influence follicular development and reproductive outcomes in clinical settings. In addition to FN1, we found that other DEGs, such as NPTX2 and CCN1, also showed significant changes in PCOS GCs. NPTX2 (neuronal pentraxin 2) is a multifunctional protein involved in synaptic plasticity and cognitive regulation in neurological diseases, as well as influencing tumor progression through epigenetic and signaling pathways. NPTX2 is significantly decreased in the cerebrospinal fluid of patients with Alzheimer's disease, and alterations in its levels have been associated with cognitive decline, synaptic loss, and tau protein pathology [ 40 ]. Mouse models have demonstrated that NPTX2 knockout in the hippocampus causes increased anxiety-like behavior, while overexpression reduces stress-induced anxiety [ 41 ]. Additionally, NPTX2 has a dual role in tumor development; for example, it is overexpressed in epithelial ovarian cancer, where it promotes tumor growth, invasion, and the hypoxia-driven malignant phenotype by activating the IL-6-JAK2/STAT3 pathway. In pancreatic cancer, however, hypermethylation of the NPTX2 promoter leads to gene silencing, and low NPTX2 expression is associated with increased tumor invasiveness, suggesting a potential role for NPTX2 as a tumor suppressor gene [ 42 ]. The role of NPTX2 may vary depending on the type of disease, cell subtype, and microenvironment. The role of NPTX2 in PCOS remains unclear and needs further investigation. CCN1 (CYR61) is an essential member of the CCN family, which includes four functional modules (IGFBP, vWFC, TSP-1, and CT). These modules regulate cell adhesion, migration, angiogenesis, and inflammation through integrin receptors. CCN1 promotes metastasis by activating the EGFR/extracellular signal-regulated kinase (ERK) pathway in gastric and breast cancers but inhibits epithelial-mesenchymal transition by blocking the transforming growth factor-β (TGF-β)/Smad pathway in pancreatic cancer [ 43 ]. In diabetic nephropathy, CCN1 promotes podocyte damage and basement membrane thickening through TGF-β-dependent pathways [ 44 ]. It has also become increasingly recognized that CCN1 plays a role in reproductive system diseases. Research shows that CCN1 levels are elevated in ectopic endometrium, and its expression is affected by estrogen. It has been observed that CCN1 encourages angiogenesis and inflammation within lesions, operating through the HIF-1α-dependent pathway. These findings suggest that CCN1 may play a key role in the connection between estrogen, angiogenesis, and inflammation [ 45 ]. Considering the pathological features of PCOS, abnormal CCN1 expression might contribute to the disease process through multiple mechanisms. Further research is needed to confirm the relationship between CCN1 and the primary characteristics of PCOS and to investigate its potential as a therapeutic target. The results showed that specific genes related to the PI3K-Akt signaling pathway were expressed at different levels. This indicates that this pathway has a significant role in the development of PCOS. PI3K-Akt is a key pathway that regulates cell growth, proliferation, and survival, and its malfunction is closely associated with various reproductive system diseases, including PCOS. Previous research has shown that abnormal activation of this pathway can cause insulin resistance and follicular development problems in patients with PCOS [ 46 , 47 ]. Our data suggest that increased expression of this pathway gene may worsen metabolic disorders and reproductive issues linked to PCOS, indicating that targeting this pathway could be a potential treatment strategy. The reduction in Akt phosphorylation following FN1 knockdown in our experiments is consistent with a functional association between FN1 and PI3K-Akt signaling activity in GCs; however, this observation alone does not establish a direct regulatory relationship. Rescue experiments employing FN1 overexpression constructs or pharmacological Akt activation will be required to determine the directionality and specificity of this interaction. This observation indicates that FN1 may influence PI3K-Akt signaling, though the precise molecular mechanism requires further investigation. The elevated FN1 expression in PCOS GCs identified in our transcriptome analysis may be associated with altered PI3K-Akt pathway activity, potentially contributing to the proliferative dysregulation characteristic of this condition; formal validation of this hypothesis through gain-of-function and pathway-specific inhibition experiments remains a priority for future investigation. Given the role of PI3K-Akt signaling in both cell proliferation and metabolic regulation, the relationship between FN1 and this pathway warrants further examination in the context of PCOS pathophysiology. Notably, the TGF-β signaling pathway was also significantly enriched in this study. TGF-β plays a crucial role in ovarian development and diseases by regulating cell growth, differentiation, and apoptosis. Imbalances in TGF-β are strongly associated with reproductive disorders like PCOS, premature ovarian failure, and ovarian cancer [ 48 ]. In PCOS pathology, abnormalities in TGF-β signaling may worsen disorders of the endocrine microenvironment by disrupting follicular development processes [ 49 ]. The TGF-β-related genes identified in this study provide new insights into the regulatory mechanisms of ovarian function and the development of PCOS. We also found that genes linked to the focal adhesion pathway were significantly enriched among the DEGs, as this pathway is essential for cell adhesion, migration, and signal transduction. Focal adhesion dysfunction can disrupt cellular responses [ 50 ]. In PCOS patients, changes in ovarian cell adhesion properties may affect follicular development and lead to ovulation issues. Identifying these genes provides new insights into the cellular and molecular mechanisms of PCOS, pointing to potential pathways for developing treatments that restore normal ovarian function. In summary, this study emphasizes the combined roles of the PI3K-Akt, TGF-β, and focal adhesion signaling pathways in the development of PCOS. These results not only enhance our understanding of the molecular basis of PCOS but also lay the groundwork for designing targeted therapies to address metabolic issues and improve reproductive outcomes. Future research will aim to confirm the roles of these key pathway components and develop innovative diagnostic methods for PCOS. The method used in this study has several advantages that ensure the reliability and scientific validity of the research findings. By collecting GC samples from both PCOS patients and non-PCOS control s simultaneously, we established a rigorous control analysis system to verify the accuracy of the high-throughput sequencing data. Follow-up multidimensional bioinformatics analysis, based on differential expression analysis, GO, KEGG, and GSEA, systematically explained the biological functional network of DEGs and further uncovered the molecular pathological mechanism of PCOS. High-expression differential genes were selected explicitly for qPCR verification, further enhancing the reliability of the data and enabling precise analysis of gene expression patterns. The biological functions and signaling pathway networks associated with the DEGs were thoroughly elucidated, providing an in-depth understanding of the molecular mechanisms underlying PCOS. We systematically explored the biological functions and signaling pathways of the DEGs, thereby deepening our understanding of the molecular mechanisms underlying PCOS. Focusing on key genes for functional experiments related to cell proliferation and cell cycle regulation emphasizes the translational medical significance of this study. It offers a theoretical foundation for developing targeted therapeutic strategies. This multi-level research approach, which combines omics analysis, molecular verification, and functional research, represents a significant breakthrough in understanding the molecular mechanisms behind PCOS. It also paves the way for more precise diagnosis and treatment of the disease in the future. However, several limitations of this study require clarification. A fundamental constraint of this study's design is that GCs from both groups were obtained under controlled ovarian hyperstimulation protocols, and control patients were infertile women with male or tubal factor infertility rather than reproductively healthy volunteers. This design, while ethically and practically necessary in the IVF/ICSI setting, means that the transcriptomic differences identified herein reflect a comparison between two infertile cohorts rather than between PCOS and normal ovarian physiology. Gn stimulation is known to substantially remodel GC gene expression, potentially confounding the identification of PCOS-specific transcriptomic signatures. Future studies should, where ethically feasible, incorporate GC samples obtained under natural cycle conditions or with minimal stimulation to enable a more physiologically authentic comparison. Firstly, the small sample size (only five patients per group) may not accurately reflect the highly heterogeneous nature of the PCOS population. This may influence the statistical power and the clinical applicability of the study's conclusions. Additionally, although high-throughput sequencing and bioinformatics analysis provide technical benefits, screening for DEGs and their functional enrichment can be easily influenced by the algorithm threshold settings. It should be emphasized that in vitro functional validation was conducted exclusively in the KGN cell line, which, as a GC tumor-derived model, harbors a heterozygous PIK3CA gain-of-function mutation and constitutively elevated PI3K-Akt activity. This characteristic may amplify FN1-dependent effects on Akt phosphorylation and apoptotic susceptibility relative to primary GCs. Accordingly, the observed proliferative and apoptotic responses following FN1 depletion should be interpreted as indicative rather than definitive, and confirmation in primary GCs isolated from both PCOS patients and matched non-PCOS infertile controls is required before translational conclusions can be drawn. Furthermore, the focus on FN1 highlights its role in cell proliferation and cycle regulation but overlooks the complexity of multigene pathway network interactions involved in PCOS pathology. Future research should increase the sample size for subtype stratification confirmation, incorporate comparative experiments using primary GCs and KGN cells, and utilize multi-omics integration to analyze the gene interaction network, thereby systematically clarifying the pathogenic mechanisms of PCOS. In conclusion, 228 DEGs were identified through transcriptome sequencing of GCs from PCOS patients and non-PCOS control s in Xinjiang, China. These genes were enriched in PCOS-related signaling pathways, such as the PI3K-Akt and MAPK pathways. qPCR confirmed that key genes, including FN1, showed abnormal expression in the GCs of PCOS patients. In vitro functional studies using the KGN cell model demonstrated that knockdown of FN1, a central key gene, significantly reduced cell proliferation and halted cell cycle progression. This indicates that FN1 may be involved in the pathological process of PCOS by regulating GC proliferation. These findings offer essential data for developing a molecular atlas of GCs in PCOS in Xinjiang, which will support further research into region-specific disease mechanisms. This study identified 228 DEGs in GCs from patients with PCOS compared to non-PCOS controls through transcriptome sequencing. Bioinformatics analysis revealed the enrichment of these genes in pathways relevant to PCOS pathophysiology, including the PI3K-Akt and MAPK signaling pathways. Among the DEGs, FN1 was significantly upregulated in PCOS GCs and validated by qPCR. In vitro functional studies demonstrated that FN1 knockdown reduced GC proliferation, induced apoptosis, and altered cell cycle distribution in the KGN cell model. Mechanistic analysis showed that FN1 depletion decreased Akt phosphorylation and modulated expression of cell cycle regulatory proteins, including cyclin D1, p21, and p27. These findings suggest that FN1 expression is associated with GC proliferative capacity and survival signaling in the context of PCOS, though the mechanistic basis of this association requires further validation through primary cell studies, rescue experiments, and in vivo models. However, several limitations should be noted. The small sample size and reliance on an immortalized cell line limit the generalizability of these findings. The clinical relevance of FN1 as a therapeutic target requires validation through primary cell studies, animal models, and clinical investigations.

Supplementary data including one table and five figures can be found with this article online at https://doi.org/10.4196/kjpp.25.418