POU2F1 drives cholesterol biosynthesis and lipid remodeling through the DHCR24/ELOVL2 axis in endometrial cancer

Abstract

Endometrial cancer (EC) is characterized by metabolic reprogramming, with cholesterol biosynthesis playing a critical role. However, the upstream transcriptional regulation of this process requires further elucidation. In this study, we identified POU2F1 as a key oncogenic transcription factor that drives cholesterol biosynthesis and tumor progression in EC. Integrative bioinformatics and clinical analyses revealed that POU2F1 is upregulated in EC and predicts poor prognosis. Mechanistically, POU2F1 directly activates DHCR24 and ELOVL2 transcription, thereby promoting DHCR24-mediated cholesterol biosynthesis and ELOVL2-associated lipid metabolic remodeling in EC. Functional assays demonstrated that POU2F1 promotes proliferation, migration, invasion, and xenograft growth in a DHCR24/ELOVL2-dependent manner. Clinically, POU2F1 expression was positively correlated with DHCR24 and ELOVL2 levels and served as an independent prognostic factor. Together, these findings establish the POU2F1-DHCR24/ELOVL2 axis as a critical driver of cholesterol-associated lipid metabolic reprogramming and cancer progression in EC, highlighting a potential therapeutic target for metabolic intervention.

Supplementary Information

The online version contains supplementary material available at 10.1186/s40170-026-00430-9.

Introduction

Endometrial cancer (EC) has become the most frequently diagnosed gynecologic malignancy in developed countries, with a steadily increasing incidence attributed to increasing obesity rates and aging populations [1]. As reported in recent global cancer statistics, in 2022, EC accounted for 420,242 new cases and 97,704 deaths worldwide [2]. Despite generally favorable prognoses for early-stage disease, patients with advanced and recurrent EC have limited treatment options and poor survival rates, with the 5-year survival rate decreasing below 20% for patients with metastatic disease. The growing prevalence of obesity-related metabolic disturbances, including dysregulated cholesterol homeostasis, has prompted increasing interest in understanding how lipid metabolic reprogramming contributes to the initiation and progression of EC.

Cholesterol, an essential constituent of cellular membranes, functions both as a precursor for steroid hormone synthesis and as a regulator of oncogenic signaling pathways [3]. Cancer cells, including those in EC, undergo a profound metabolic rewiring to sustain uncontrolled proliferation, survival under stress, and metastatic dissemination [4]. This reprogramming often involves heightened demands for essential macromolecular building blocks, among which cholesterol stands out as crucial for maintaining membrane integrity, facilitating signal transduction (e.g., through lipid rafts), and serving as a precursor for bioactive molecules. Key enzymes in the mevalonate pathway—such as 3-hydroxy-3-methylglutaryl-CoA synthase 1 (HMGCS1), acetyl-CoA acetyltransferase 2 (ACAT2), and 24-dehydrocholesterol reductase (DHCR24)—are frequently overexpressed in EC and are correlated with advanced disease [5–7]. DHCR24, which catalyzes the final step in cholesterol biosynthesis, not only maintains membrane integrity but also protects cancer cells from oxidative stress and apoptosis [8]. Additionally, ELOVL fatty acid elongase 2 (ELOVL2), which regulates the production of very long-chain fatty acids (VLCFAs), influences membrane fluidity and lipid raft formation, further enhancing cancer progression [9, 10]. These findings underscore the importance of cholesterol and fatty acid metabolism in EC progression; however, the upstream transcriptional mechanisms controlling these pathways remain poorly understood.

The POU-domain transcription factor POU class 2 homeobox 1 (POU2F1, also known as OCT1) is a versatile regulator of gene expression involved in development, immunity, and cellular metabolism [11]. In cancer, POU2F1 often acts as an oncogene by promoting proliferation, stemness, immune escape, and metastasis [11–13]. It has been shown to regulate metabolic reprogramming in colon cancer by modulating the expression of genes related to glucose metabolism [11]. Regrettably, the role of POU2F1 in EC progression, particularly its oncogenic mechanisms related to cholesterol metabolism, remains poorly characterized. Given its known function as a transcriptional regulator of metabolic genes, we hypothesized that POU2F1 may directly modulate DHCR24 and ELOVL2, thereby promoting DHCR24-mediated cholesterol biosynthesis and ELOVL2-associated lipid remodeling to drive EC pathogenesis.

This study aims to elucidate the mechanistic link between POU2F1, cholesterol-associated lipid metabolic reprogramming, and EC malignancy by focusing on two key metabolic enzymes, DHCR24 and ELOVL2. We propose that POU2F1 transcriptionally upregulates these enzymes, thereby enhancing DHCR24-mediated cholesterol biosynthesis and ELOVL2-associated lipid remodeling to promote cancer progression. Our findings identify POU2F1 as an important regulator of cholesterol-associated lipid metabolic reprogramming in EC and identify DHCR24 and ELOVL2 as potential therapeutic targets, offering new strategies to disrupt cholesterol-dependent tumor growth and improve outcomes for high-risk patients.

Materials and methods

Bioinformatics analysis

Clinical and genomic data from 35 normal endometrial tissue samples and 554 EC samples were obtained from The Cancer Genome Atlas Uterine Corpus Endometrial Carcinoma (TCGA-UCEC) database and processed as previously described [14]. Transcription factors (TFs) and binding sites were predicted using the animal TFDB 4.0 database [15] and the JASPAR database [16].

Cell culture and reagents

The human cancer cell lines HEC251, AN3CA, HEC-6, Ishikawa, and KLE, along with human endometrial stromal cells (ESCs), were obtained from the China Center for Type Culture Collection (Hubei, China). All cells were maintained at 37 °C in a humidified atmosphere containing 5% CO₂, following the supplier’s recommended culture conditions. Additionally, simvastatin (HY-17502) and mevalonic acid (MVA) lithium salt (HY-113071 A) were purchased from MedChemExpress. Mevalonic acid lithium salt (hereafter referred to as mevalonate) [17] was used to rescue the inhibition of the mevalonate pathway induced by simvastatin.

CCK-8 and EdU assays

The cell viability and proliferative capacity of cancer cells were evaluated using CCK-8 and EdU assays, which were performed in accordance with previously published protocols [14, 18–22]. For the CCK-8 assay, cells were seeded into 96-well plates at a density of 2 × 10³ cells/well in 100 µL complete medium and cultured overnight to allow attachment. At the indicated time points (0, 24, 48, 72, and 96 h), 10 µL CCK-8 reagent (G4103, Servicebio) was added to each well, followed by incubation at 37 °C for 1–2 h. The absorbance was then measured at 450 nm using a microplate reader. Each experimental group contained at least five replicate wells, and all experiments were performed independently at least three times.

For the EdU assay, cells were seeded into 24-well plates containing coverslips at a density of 5 × 10⁴ cells/well and cultured overnight. Cells were then incubated with EdU (G1602, Servicebio) working solution (10 µM) for 2 h at 37 °C according to the manufacturer’s instructions. After incubation, cells were fixed with 4% paraformaldehyde for 15 min, permeabilized with 0.3% Triton X-100 for 15 min, and stained using the EdU detection kit. Nuclei were counterstained with DAPI. Images were captured under a fluorescence microscope, and the percentage of EdU-positive cells was calculated from at least five randomly selected fields for each group. All experiments were repeated at least three times.

Migration and invasion assays

Overall, the migration and invasion assays were performed as previously described [14, 18–22]. Cell migration and invasion were assessed using Transwell chambers with 8-µm pore size inserts (#3422, Corning). For the migration assays, 5 × 10⁴ cells were resuspended in serum-free medium and seeded into the upper chambers, while medium supplemented with 10% fetal bovine serum was added to the lower chambers. For the invasion assays, the upper chambers were precoated with Matrigel according to the manufacturer’s instructions prior to cell seeding.

The migration and invasion assays were conducted for 24 h, after which nonmigrated or noninvaded cells on the upper surface of the membrane were gently removed. Cells that had migrated to or invaded the lower surface were fixed, stained, and imaged under a light microscope. Quantification was performed by counting cells in multiple randomly selected fields.

Intracellular free cholesterol, total cholesterol, and neutral lipid assays

The assessments of free cholesterol levels, total cholesterol levels, and intracellular lipid droplets were performed using a Free Cholesterol Kit (BC1895, Solarbio), a Total Cholesterol Kit (BC1985, Solarbio), and a BODIPY 493/503 Lipid Droplet Kit (C2053S, Beyotime), respectively, following the manufacturers’ recommendations and previously reported protocols [20].

Reverse transcription quantitative polymerase chain reaction (RT‒qPCR)

In short, RT‒qPCR was performed following previously published protocols [14, 18–21]. Total RNA was extracted from cultured cells and tissue samples using a commercial Total RNA Extraction Kit (R1200; Solarbio) in strict accordance with the manufacturer’s instructions. RNA concentration and purity were assessed by measuring the absorbance at 260 and 280 nm using an SMA2000 spectrophotometer (Thermo Fisher Scientific, Inc.); samples with A260/A280 ratios ranging from 1.8 to 2.0 were considered suitable for subsequent analyses. Complementary DNA (cDNA) was synthesized from total RNA using a reverse transcription system following the supplier’s protocol (13089ES, Yeasen). Quantitative real-time PCR (RT‒qPCR) was performed under the following cycling conditions: initial denaturation at 94 °C for 5 min, followed by 40 amplification cycles consisting of denaturation at 94 °C for 20 s and annealing at 60 °C for 20 s. β-actin was used as an internal reference gene to normalize target gene expression. Relative mRNA expression levels were calculated using the 2⁻ΔΔCq method. All reactions were performed in triplicate to ensure reproducibility. The sequences of the primers used for RT‒qPCR are listed in Table S1.

Western blotting

Western blotting was performed in accordance with established procedures [14, 18–21]. Cells and tissue samples were lysed using radioimmunoprecipitation assay (RIPA) buffer supplemented with protease inhibitors. Protein concentrations were determined with a bicinchoninic acid (BCA) protein assay kit (P0012; Beyotime) according to the manufacturer’s instructions. Equal amounts of protein (20 µg per sample) were resolved by 10% SDS–PAGE and subsequently transferred onto polyvinylidene difluoride (PVDF) membranes (Millipore). The membranes were blocked with 5% nonfat milk at room temperature for 2 h and then incubated with primary antibodies against POU2F1 (1:1000, TB4307, Abmart), DHCR24 (1:1000, 10471-1-AP, Proteintech), ELOVL2 (1:1000, 20308-1-AP, Proteintech), ELOVL3 (1:1000, 32474-1-AP, Proteintech), GLDC (1:1000, PK47535, Abmart), TRIB3 (1:2000, 13300-1-AP, Proteintech), β-actin (1:2000, GB11001-100, Servicebio), GAPDH (1:1000, GB11002-100, Servicebio) and Tubulin (1:2000, GB11017-100, Servicebio). After incubation with the appropriate horseradish peroxidase–conjugated secondary antibody (goat anti-rabbit IgG; 1:1000; SE134; Solarbio) for 1 h at room temperature, immunoreactive bands were visualized using an enhanced chemiluminescence detection system (Thermo Fisher Scientific, Inc.). Band intensities were quantified using ImageJ software (version 1.8.0), with GAPDH, β-actin, or tubulin serving as internal loading controls.

Gene overexpression and knockdown

Human POU2F1 (Gene ID: 5451), DHCR24 (Gene ID: 1718), and ELOVL2 (Gene ID: 54898) overexpression vectors, as well as gene-specific and control shRNA plasmids, were acquired from VectorBuilder (Guangzhou, China). The sequences of the shRNAs are presented as follows: POU2F1 sh-RNA#1: GTACAGTCTAAATCCAGTGAA; POU2F1 sh-RNA#2: GCTGCTCAGTCTTTAAATGTA; DHCR24 sh-RNA#1: CGAGTCATCATCCCACAAGTA; DHCR24 sh-RNA#2: CCAACACATCTGCACTGCTTA, ELOVL2 sh-RNA#1: TATGTTTGGACCGCGAGATTC; and ELOVL2 sh-RNA#2: GGTGCTTTGGTGGTACTATTT. To achieve simultaneous manipulation of the two genes, lentiviral vectors carrying different antibiotic resistance cassettes were used. Specifically, puromycin resistance was incorporated into the overexpression constructs, whereas geneticin (G418) resistance was included in the knockdown vectors. Following lentiviral transduction, the cells were subjected to antibiotic selection using puromycin (2 µg/mL; P8230; Solarbio) for overexpression or G418 (400 µg/mL; IG0010, Solarbio) for knockdown. Selection was maintained for 7–10 days until the complete elimination of nontransduced control cells was confirmed. The efficiency of gene overexpression and knockdown was subsequently validated by RT‒qPCR and Western blot analyses [18, 20].

Dual-luciferase reporter, ChIP and qPCR assays

To evaluate the transcriptional activity of the DHCR24 and ELOVL2 promoters, genomic DNA fragments encompassing the promoter regions of DHCR24 (− 2000 to + 99) and ELOVL2 (− 2000 to + 99, relative to the transcription start site) were amplified, subcloned and inserted into the pGL6-dura luciferase reporter vector (VT008344, Solarbio). All the constructs were verified by DNA sequencing prior to use. Cells were cotransfected with the indicated luciferase reporter plasmids and expression vectors using standard transfection reagents. A Renilla luciferase plasmid was included as an internal control to normalize transfection efficiency. After 24–48 h of transfection, luciferase activity was measured using a dual-luciferase reporter assay system according to the manufacturer’s instructions and as previously described [14, 18–21]. Firefly luciferase activity was normalized to Renilla luciferase activity. The following primer sequences were applied for the DHCR24 promoter region (− 665/−511) and the ELOVL2 region (− 2031/−1880): DHCR24-forward (5’-GTACACAATGGAGCTCACCACT-3′), DHCR24-reverse (5′-GATGGAAACCTGGCGGTAACCT-3′), and ELOVL2-forward (5′-CGCAAACCTGCAGGAACAGAGC-3′), and ELOVL2-reverse (5′-CCAGGAATACCCACCCGCTG-3′).

ChIP assays were performed using a ChIP kit (P2078, Beyotime) following established protocol [14, 18–21]. Briefly, cells were cross-linked with 1% formaldehyde, quenched with glycine, and lysed. Chromatin was sonicated to generate DNA fragments of approximately 200–500 bp. Immunoprecipitation was performed using antibodies against POU2F1 or control IgG. After the reversal of cross-links and DNA purification, enriched DNA fragments were quantified by qPCR. To map POU2F1 binding sites within the DHCR24 promoter, three distinct regions were analyzed: DHCR24-E1-forward (− 1991/−1904) (5′-GACTCCATCCTTCTCTTCCC-3′), DHCR24-E1-reverse (− 1991/−1904) (5′-ATTAACTGAAGAGTGTTGTGTAGG-3′), DHCR24-E2-forward (-1938/-1798) (5′-CCACTGTTTCTCCTACACAACACT-3′), DHCR24-E2-reverse (-1938/-1798) (5′-CCCCGATTCTGTCTGACTCTGTTA-3′), DHCR24-E3-forward (-956/-801) (5′-GCCAGACAGAGAAGAACCTGGTAT-3′), DHCR24-E3-reverse (-956/-801) (5′-TGCCCTCACTCAGCAACCAG-3′), ELOVL2-forward (-1621/-1460) (5′-ACCCATATTAATTTAACCCTAACAGC-3′), and ELOVL2-reverse (-1621/-1460) (5′-CCTAGTCTGATAAAAATCTCAGGAGT-3′). ChIP-qPCR results were calculated as the fold enrichment relative to that of the IgG controls and input DNA.

Immunofluorescence (IF), HE staining and immunohistochemistry (IHC)

The experimental procedures for IF, HE, and IHC were performed as previously described [14, 18–21].

For HE staining, paraffin-embedded tissue sections were deparaffinized in xylene, rehydrated through graded ethanol solutions, stained with hematoxylin and eosin, dehydrated and subsequently mounted for histopathological evaluation.

For IHC analysis, paraffin-embedded tissue Sect. (4 μm) were deparaffinized, rehydrated, and subjected to antigen retrieval using citrate buffer (pH 6.0). Endogenous peroxidase activity was blocked with 3% hydrogen peroxide, followed by incubation with blocking solution to reduce nonspecific binding. The sections were then incubated overnight at 4 °C with the following primary antibodies: anti-POU2F1 (1:100, TB4307, Abmart), anti-DHCR24 (1:200, 10471-1-AP, Proteintech), anti-ELOVL2 (1:100, 20308-1-AP, Proteintech), and anti-Ki-67 (1:100, GB111499, Servicebio). After the sections were washed, they were incubated with appropriate horseradish peroxidase–conjugated secondary antibodies, and immunoreactivity was visualized using diaminobenzidine (DAB) substrate. The nuclei were counterstained with hematoxylin, and the slides were dehydrated and mounted. The proliferation index was determined by calculating the percentage of Ki-67-positive nuclei relative to the total number of nuclei in representative high-power fields. Quantitative analysis of IHC staining for POU2F1, DHCR24, and ELOVL2 was performed using ImageJ software by calculating the average optical density (AOD), defined as integrated optical density (IOD) divided by the measured area [14, 18–21].

IF staining was performed in cultured cells following standard procedures as previously described [14, 17–20]. Cells were seeded onto sterile glass coverslips and allowed to adhere overnight. After being washed with phosphate-buffered saline (PBS), the cells were fixed with 4% paraformaldehyde at room temperature for 15 min, permeabilized with 0.1% Triton X-100 for 10 min, and blocked with 5% bovine serum albumin (BSA) for 1 h to reduce nonspecific binding. The cells were then incubated with anti-POU2F1 primary antibodies (1:50, TB4307, Abmart) for 2 h at room temperature. After being washed with PBS, the cells were incubated with a Cy3-conjugated goat anti-rabbit IgG secondary antibody (1:50; GB21303; Servicebio) at room temperature for 1 h in the dark. The nuclei were counterstained with DAPI, and the coverslips were mounted with anti-fade mounting medium. Fluorescence images were captured using a fluorescence microscope under identical imaging conditions for all groups.

Animal assays

The animal experiments were conducted in accordance with the ARRIVE guidelines and approved by the Ethics Committee of Xiangyang No. 1 People’s Hospital. For the in vivo studies, five-week-old female BALB/c nude mice were used for xenograft tumor growth and pulmonary metastasis assays (n = 5 per group). In the xenograft model, 1 × 10⁶ stably transfected cancer cells were subcutaneously inoculated into the right dorsal flank of each mouse. Tumor progression was monitored twice weekly using caliper measurements, and tumor volume was calculated as follows: volume = length×(width)²/2. The experiment was terminated if any tumor reached approximately 1000 mm³ [23], at which point the mice were humanely euthanized. For the lung metastasis experiment, 5 × 10⁶ stably transfected cells were injected into mice through the tail vein. After 35 days, the animals were euthanized, and metastatic lesions in the lungs were examined by H&E staining.

Human tissues

All enrolled patients were treatment-naïve prior to surgery, with EC confirmed by routine pathological examination. This study protocol was approved by the Ethics Committee of Xiangyang No. 1 People’s Hospital. In total, we collected 30 paired EC and adjacent normal endometrial tissue samples for RT‒qPCR analysis to evaluate the mRNA expression levels of POU2F1, DHCR24, and ELOVL2. Among these, 15 matched pairs were further subjected to Western blotting to assess protein expression. Additionally, IHC was performed to measure the protein levels of POU2F1, DHCR24, and ELOVL2 in 187 patients with EC patients and their matched adjacent endometrial tissues. IHC scoring was conducted as previously described [14, 18–21]. Pearson correlation analysis was used to assess the relationships between POU2F1 and its target genes (DHCR24/ELOVL2).

Statistical analysis

All experiments were repeated in triplicate, and the data are presented as the mean ± standard deviation. For comparisons between two independent groups, a two-tailed Student’s t test was employed, while paired observations were evaluated using a paired t test. Multigroup comparisons were conducted through one-way ANOVA with Bonferroni-adjusted post hoc tests. Survival curves were assessed by log-rank tests. A significance threshold of p < 0.05 was applied for all tests.

Results

POU2F1 is a key TF involved in cholesterol metabolism and malignant progression of EC

Given the critical role of cholesterol metabolism in EC, we conducted a comprehensive analysis of key cholesterol metabolism genes using the TCGA-UCEC dataset. Our findings revealed differential expression patterns of 105, 83, 32, and 7 cholesterol metabolism genes in tumor versus normal tissues, G3 versus G1-2 grades, stages III-IV versus stages I-II, and dead versus alive patient groups, respectively (Fig. 1A). Through intersection analysis, we identified five cholesterol metabolism genes (DHCR24, ELOVL2, ELOVL3, GLDC and TRIB3) that were consistently upregulated in tumor tissues, high-grade tumors, advanced-stage cancers, and deceased patients (Fig. 1A, Fig. S1). These genes were significantly associated with poor clinical outcomes (Fig. S2A-E). Additionally, RT-qPCR and western blot analyses confirmed that the expression levels of these five-cholesterol metabolism-related genes were significantly increased in tumor cell lines (Fig. S2F-G). Furthermore, we performed an integrated analysis of key TFs using the TCGA-UCEC dataset. Comparative analyses between tumor versus normal tissues, G3 versus G1-2 grades, stages III-IV versus stages I-II, and dead versus alive patients revealed 1,145, 969, 421, and 55 differentially expressed TFs, respectively (Fig. 1A). Through overlap analysis, we discovered 10 TFs that were consistently dysregulated in tumor tissues, high-grade tumors, advanced-stage disease, and deceased patients. Intersection analysis of these 10 differentially expressed TFs with 216 cholesterol metabolism-related TFs predicted by the AnimalTFDB 4.0 database [15] revealed 4 overlapping regulatory factors (E2F1, ESR1, PGR, and POU2F1) that may modulate cholesterol metabolism in EC (Fig. 1A). Among these genes, E2F1 and POU2F1 were consistently upregulated in tumor tissues, high-grade tumors, advanced-stage disease, and deceased patients, with elevated expression levels significantly correlated with poor prognosis (Fig. 1B–C, Fig. S3A-B). Conversely, ESR1 and PGR were persistently downregulated in these same high-risk groups, and their reduced expression was strongly associated with unfavorable clinical outcomes (Fig. S3A-B). Further experiments demonstrated that POU2F1 and E2F1 were upregulated, whereas ESR1 and PGR were downregulated in EC cell lines (Fig. S3C-E). Research has shown that E2F1 promotes EC malignancy by upregulating the expressions of trophinin-associated protein (TROAP) and BMI1 [24, 25]. ER overexpression and ESR1 amplification occur early in endometrial tumorigenesis, but ESR1 expression decreases with tumor progression [26, 27]. However, our data indicated that ESR1 expression is reduced in high-risk and advanced EC, suggesting that ER signaling is dynamically regulated during disease progression [26]. In addition, the loss of PGR expression in EC is connected to unfavorable patient prognosis [28]. Accumulating evidence indicates that POU2F1 contributes to metabolic reprogramming in multiple cancer types, including colon cancer [11], and gastric cancer [13], and drives malignant progression. Nevertheless, the role of POU2F1 in EC pathogenesis has yet to be investigated. The lack of studies on POU2F1 in EC, combined with bioinformatics predictions indicating its tumorigenic potential and correlation with unfavorable outcomes, prompted us to select POU2F1 as a candidate TF for further research. Additionally, the UALCAN database [29] revealed significantly higher POU2F1 protein expression levels in EC tissues than in normal endometrial tissues (Fig. 1D). To further validate the expression pattern of POU2F1 in EC, we performed RT‒qPCR and Western blot analyses in five EC cell lines and normal endometrial stromal cells. Our results demonstrated consistent upregulation of POU2F1 across multiple EC cell lines, with the highest expression level observed in KLE cells (Fig. 1E-F). Considering the distinct POU2F1 expression patterns observed in EC cell lines, we chose the KLE cell line (highest POU2F1 expression) and Ishikawa cell line (lowest POU2F1 expression) for further mechanistic studies. Our results indicated that stable POU2F1 overexpression markedly increased DHCR24 and ELOVL2 expression at both the mRNA and protein levels in Ishikawa cells (Fig. 1G and I). Conversely, stable knockdown of POU2F1 markedly suppressed DHCR24 and ELOVL2 expression at both the transcriptional and translational levels in KLE cells (Fig. 1H and I). Notably, the expression patterns of three other candidate genes (ELOVL3, GLDC, and TRIB3) remained unaffected by POU2F1 manipulation (Fig. 1G-I). Our data consistently indicated that POU2F1 overexpression, which is associated with unfavorable clinical outcomes, may promote EC by enhancing cholesterol biosynthesis, potentially through regulation of the key metabolic enzymes DHCR24 and ELOVL2.

Fig. 1.

POU2F1 is a key TF in cholesterol metabolism in EC. (A) The Venn diagram illustrates a comprehensive analysis of key cholesterol metabolism genes (left panel) and potential TFs (right panel) in EC, based on the TCGA-UCEC dataset and the AnimalTFDB 4.0 database (FDR < 0.05). (B) The differential expression of POU2F1 across various subgroups of EC based on the TCGA-UCEC dataset. (C) The survival curve illustrates the differences in overall survival of POU2F1 based on the TCGA-UCEC dataset. (D) The protein expression levels of POU2F1 between the normal and EC groups derived from the UALCAN database. (E-F) The expression level of POU2F1 between normal endometrial stromal cells and five EC cell lines detected by RT-qPCR (E, n = 5) and western blot (F, n = 5), respectively. (G-H) The mRNA levels of POU2F1 and five potential target genes (DHCR24, ELOVL2, ELOVL3, GLDC, and TRIB3) were measured by RT-qPCR in Ishikawa (G, n = 3) and KLE (H, n = 5) cells that were stably transfected as indicated, respectively. (I) The protein levels of POU2F1 and five potential target genes (DHCR24, ELOVL2, ELOVL3, GLDC, and TRIB3) were measured by western blot in Ishikawa and KLE cells that were stably transfected as indicated, respectively (n = 5).*P < 0.05, **P < 0.01, ***P < 0.001

POU2F1 promotes DHCR24-mediated cholesterol biosynthesis and ELOVL2-associated lipid remodeling in EC

To further elucidate the effect of POU2F1 on cholesterol biosynthesis in EC, we examined alterations in total cholesterol, free cholesterol, and lipid droplet content in EC cell lines (Ishikawa and KLE cells) following stable overexpression or knockdown of POU2F1. These data indicated that stable POU2F1 overexpression increased the levels of total cholesterol, free cholesterol, and intracellular lipid droplets in Ishikawa cells (Fig. 2A-B and E). Conversely, stable POU2F1 knockdown markedly reduced these lipid parameters in KLE cells. These findings demonstrate that POU2F1 is involved in cholesterol biosynthesis in EC (Fig. 2C-D F).

Fig. 2.

POU2F1 promotes cholesterol metabolism in EC. (A-D) The bar charts show the levels of total cholesterol and free cholesterol in Ishikawa (A-B, n = 3) and KLE (C-D, n = 5) cells were stably transfected as indicated, respectively. (E-F) The content of neutral lipid was assessed by BODIPY 493/503 assay in Ishikawa (E, n = 3) and KLE (F, n = 5) cells stably transfected as indicated, respectively. (G-I) The total cholesterol (G, n = 5) and free cholesterol (H, n = 5) and the content of neutral lipids (I, n = 5) in Ishikawa cells were stably transfected as indicated, respectively. (J-L) The total cholesterol (J, n = 5) and free cholesterol (K, n = 5) and the content of neutral lipids (L, n = 5) in KLE cells were stably transfected as indicated, respectively. (M-N) Western blot assays were used to evaluate the protein levels of POU2F1, DHCR24, and ELOVL2 in Ishikawa (M, n = 3) and KLE (N, n = 3) cells stably transfected as indicated, respectively. *P < 0.05, **P < 0.01, ***P < 0.001

To evaluate whether the phenotypic effects associated with POU2F1 overexpression are dependent on the mevalonate-cholesterol synthesis pathway, Ishikawa cells were treated with simvastatin, a pharmacological inhibitor of HMG-CoA reductase [30], or were cotreated with mevalonate [17], a downstream metabolite of the pathway. Simvastatin was used to inhibit cholesterol biosynthesis, and mevalonate supplementation was used to determine whether the observed effects were specifically mediated through the mevalonate pathway. As shown in Fig. S4A-4B, simvastatin treatment significantly reduced both total cholesterol and free cholesterol levels, whereas POU2F1 overexpression markedly increased the cholesterol content. Notably, mevalonate supplementation effectively restored cholesterol levels suppressed by simvastatin, even in cells overexpressing POU2F1. Consistent with these findings, quantitative analysis of lipid droplet accumulation demonstrated that simvastatin significantly decreased lipid storage, whereas POU2F1 overexpression enhanced lipid droplet formation. This reduction was largely reversed upon mevalonate treatment (Fig. S4C-D). Collectively, these results show that pharmacological inhibition of the mevalonate pathway by simvastatin suppresses lipid accumulation and malignant phenotypes in Ishikawa cells, whereas mevalonate supplementation effectively rescues these effects, supporting a functional association between POU2F1-related phenotypes and the mevalonate-cholesterol biosynthesis axis.

To further investigate whether the POU2F1-mediated promotion of cholesterol biosynthesis depends on DHCR24 and ELOVL2, we performed additional experiments. The results revealed that stable knockdown of either DHCR24 or ELOVL2 expression significantly reduced total cholesterol, free cholesterol, and lipid droplet levels in Ishikawa cells (Fig. 2G-I). However, concomitant stable overexpression of POU2F1 reversed these inhibitory effects (Fig. 2G-I). Conversely, stable knockdown of POU2F1 reversed the stimulatory effects of DHCR24 and ELOVL2 overexpression on cholesterol biosynthesis in KLE cells (Fig. 2J–L). Taken together, our findings indicate that POU2F1 promotes cholesterol biosynthesis in EC through DHCR24-mediated cholesterol biosynthesis and ELOVL2-associated lipid remodeling.

POU2F1 promotes the EC progression in a manner dependent on DHCR24 and ELOVL2

Next, we further investigated the oncogenic role of POU2F1 in EC. Stable overexpression of POU2F1 markedly promoted the viability, proliferation, migration, and invasion of Ishikawa cells (Fig. 3A and C, and 3E). Conversely, stable POU2F1 knockdown markedly suppressed these malignant phenotypes in KLE cells (Fig. 3B and D, and 3F). We performed rescue experiments to determine whether the tumor-promoting effects of POU2F1 depend on DHCR24 and ELOVL2. Our data revealed that knockdown of either DHCR24 or ELOVL2 expression significantly inhibited cell viability, proliferation, migration, and invasion in Ishikawa cells, whereas concomitant POU2F1 overexpression reversed these suppressive effects (Fig. 4A, C and E, and 4G). Conversely, overexpression of DHCR24 or ELOVL2 enhanced the oncogenic behaviors of KLE cells, but these effects were reversed by POU2F1 knockdown (Fig. 4B, D and F, and 4H). More importantly, POU2F1 overexpression partially reversed the inhibitory effects of simvastatin on EC cell viability, proliferation, migration and invasion, and cotreatment with mevalonate further enhanced the reversal of simvastatin-induced suppression (Fig. S4E-I). Collectively, these results indicate that the protumorigenic effects of POU2F1 on EC cell proliferation, migration and invasion are closely associated with an intact mevalonate–cholesterol biosynthesis pathway. Therefore, the oncogenic functions of POU2F1 in EC are at least partly dependent on DHCR24-mediated cholesterol biosynthesis and ELOVL2-associated lipid remodeling.

Fig. 3.

POU2F1 promotes EC progression in a manner dependent on DHCR24 and ELOVL2. (A-B) CCK-8 assays evaluating the cell viability in Ishikawa (A, n = 5) and KLE (B, n = 5) cells were stably transfected as indicated, respectively. Statistical significance was calculated at the 96 h time point. (C-D) EdU assays evaluating the cell proliferation ability in Ishikawa (C, n = 5) and KLE (D, n = 5) cells were stably transfected as indicated, respectively. (E-H) Transwell migration and invasion assays evaluating the cell migration and invasion abilities in Ishikawa (E, G, n = 5) and KLE (F, H, n = 5) cells were stably transfected as indicated, respectively. Migration and invasion assays were performed for 24 h under serum-free conditions in the upper chambers. *P < 0.05, **P < 0.01, ***P < 0.001

Fig. 4.

POU2F1 promotes the malignant progression of EC in vitro. (A-B) CCK-8 assays evaluating the cell viability in Ishikawa (A, n = 3) and KLE (B, n = 5) cells were stably transfected as indicated, respectively. Statistical significance was calculated at the 96 h time point. (C-D) EdU assays evaluating the cell proliferation ability in Ishikawa (C, n = 3) and KLE (D, n = 5) cells were stably transfected as indicated, respectively. (E-F) Transwell migration and invasion assays evaluating the cell migration and invasion abilities in Ishikawa (E, n = 3) and KLE (F, n = 5) cells were stably transfected as indicated, respectively. Migration and invasion assays were performed for 24 h under serum-free conditions in the upper chambers. *P < 0.05, **P < 0.01, ***P < 0.001

POU2F1 regulates the promoter activity of the target genes DHCR24 and ELOVL2

We subsequently investigated the transcriptional regulatory effects of POU2F1 on its target genes DHCR24 and ELOVL2. First, we examined the subcellular localization of the POU2F1 protein in HEC-6 and KLE cells, which exhibit relatively high endogenous POU2F1 expression among EC cell lines (Figs. 1F and 5A–B). Immunofluorescence analysis revealed predominant nuclear localization of POU2F1 in both cell lines (Fig. 5A–B). This nuclear localization pattern further supports the notion that POU2F1 exerts its oncogenic functions through transcriptional regulation in EC cells. To further elucidate the transcriptional regulation of DHCR24 and ELOVL2 by POU2F1, we performed dual-luciferase reporter assays. The results demonstrated that stable POU2F1 overexpression significantly increased the promoter activity of both DHCR24 and ELOVL2 in Ishikawa cells, whereas POU2F1 knockdown markedly suppressed their promoter activity in KLE cells (Fig. 5C-D and J-K). To investigate the molecular mechanism underlying this regulation, we used the JASPAR database [16] to predict potential POU2F1 binding sites within the promoter regions of these target genes. Bioinformatics analysis revealed three putative POU2F1 binding sites in the DHCR24 promoter and one conserved binding motif in the ELOVL2 promoter region (Fig. 5E-F and L-M). To further validate the functional binding sites of POU2F1 on the DHCR24 and ELOVL2 promoters, we constructed both wild-type and mutant promoter reporter plasmids. Subsequent dual-luciferase reporter assays demonstrated that mutations in binding site 3 of the DHCR24 promoter significantly reduced transcriptional activity compared with that in the wild-type construct, suggesting that this cis-element is critical for the POU2F1-mediated regulation of DHCR24 expression (Fig. 5G). Similarly, mutations in the predicted binding site in the ELOVL2 promoter resulted in markedly decreased luciferase activity (Fig. 5N). To confirm binding specificity, we performed ChIP-qPCR and agarose gel electrophoresis analysis. Both methods consistently abolished POU2F1 enrichment at the mutated binding sites (Fig. 5H and O). These comprehensive findings provide strong evidence that DHCR24 and ELOVL2 are direct transcriptional targets of POU2F1 in EC cells. Notably, POU2F1 expression was significantly positively correlated with DHCR24 (R = 0.392; P < 0.001) and ELOVL2 (R = 0.358; P < 0.001) expression in the TCGA-UCEC cohort (Fig. 5I and P). In conclusion, these results consistently indicate that POU2F1 regulates the transcriptional levels of DHCR24 and ELOVL2 in EC.

Fig. 5.

POU2F1 regulates the transcriptional activity of DHCR24 and ELOVL2. (A-B) Immunofluorescence analysis showing the subcellular localization of the POU2F1 protein in HEC-6 (A) and KLE (B) cells. (C-D, J-K) Dual luciferase assay evaluating the transcriptional activity of DHCR24 (C-D) and ELOVL2 (J-K) promoter in Ishikawa and KLE cells stably transfected as indicated, respectively (n = 5). (E and L) Canonical POU2F1 binding motif based on the JASPAR database. (F and M) Prediction of potential POU2F1 binding sites on DHCR24 (F) and ELOVL2 (M) promoters based on JASPAR database. (G and N) Dual luciferase assay evaluating the transcriptional activity of wild-type or mutant DHCR24 (G) and ELOVL2 (N) promoter in Ishikawa cells stably transfected as indicated, respectively (n = 5). (H and O) ChIP-qPCR assay indicating that the POU2F1 protein was enriched in the DHCR24 (H) and ELOVL2 (O) promoter region (IgG as a negative control) (n = 3). (I) The correlation scatter plot indicating the correlation between DHCR24 and POU2F1 based on the TCGA-UCEC database. (P) The correlation scatter plot indicating the correlation between ELOVL2 and POU2F1 based on the TCGA-UCEC database. *P < 0.05, **P < 0.01, ***P < 0.001

Knocking down POU2F1 expression inhibits the malignant progression of EC in vivo

To further investigate the oncogenic function of POU2F1 in EC, we established xenograft models by subcutaneously inoculating POU2F1-knockdown EC cells (sh-POU2F1) and control cells (sh-NC) into immunodeficient mice. Longitudinal monitoring over a 5-week period revealed that compared with control treatment, POU2F1 depletion significantly attenuated tumor growth, as evidenced by reduced tumor growth rate, tumor volume, and final tumor weight (Fig. 6A–C). Western blotting and immunohistochemical analyses of excised tumors revealed marked downregulation of the expression of POU2F1 and its correlated targets DHCR24 and ELOVL2 in the sh-POU2F1(Fig. 6D–F). More importantly, experimental metastasis assays via tail vein injection revealed that POU2F1 knockdown substantially decreased pulmonary metastatic nodule formation while significantly prolonging overall survival (log-rank P < 0.05) (Fig. 6G-I). These collective findings demonstrate that POU2F1 drives EC progression in vivo by promoting tumor growth, proliferation, and metastatic dissemination, underscoring its potential as a novel therapeutic target for EC management.

Fig. 6.

Stable knockdown of POU2F1 inhibits the progression of EC in vivo. The tumor photographs (A, n = 5 per group), growth curves (B, n = 5 per group), and tumor weight (C, n = 5 per group) of subcutaneous transplanted tumors with KLE cells stably transfected as indicated, respectively (n = 5 per group). (D-F) Western blot (D, n = 3) and IHC (E-F) assays evaluating the protein levels of POU2F1, DHCR24, ELOVL2 and Ki-67 in subcutaneous transplanted tumors as indicated, respectively (n = 5 per group). (G-I) The representative HE staining of lung (G, right panel), lung metastatic nodules count (H), and survival curve (I) of the lung metastasis experiment injected with KLE cells, respectively (n = 5 per group). *P < 0.05, **P < 0.01, ***P < 0.001

Evaluation of the expression of POU2F1/DHCR24/ELOVL2 in EC tissues

We next examined the expression levels of POU2F1, DHCR24, and ELOVL2 in EC tissues. RT‒qPCR analysis of 30 paired samples revealed that all three genes were significantly upregulated at the mRNA level in EC tissues compared with adjacent normal endometrium, a finding further confirmed at the protein level by Western blotting in 15 paired cases (Fig. 7A-B). Consistent with these findings, IHC in a larger cohort of 187 patients revealed markedly increased expression of POU2F1, DHCR24, and ELOVL2 in tumor tissues compared with matched controls (Fig. 7C-F). Moreover, a correlation analysis based on the IHC results revealed that POU2F1 expression was positively associated with both DHCR24 (R = 0.332, P < 0.001) and ELOVL2 (R = 0.384, P < 0.001) expression (Fig. 7G), which is in line with the correlations observed in the TCGA-UCEC dataset (Fig. 5I and P). Collectively, these results indicate that the concurrent overexpression of POU2F1, DHCR24, and ELOVL2 is a common feature of EC and may contribute to its unfavorable clinical outcomes.

Fig. 7.

POU2F1, DHCR24, and ELOVL2 are evaluated in EC tissues. (A) RT-qPCR (A, n = 30) and western blot (B, n = 15) evaluating the mRNA and protein levels of POU2F1, DHCR24, and ELOVL2 between EC tissues and adjacent normal tissues. (C-F) IHC assays measuring protein expression of POU2F1 (C and D), DHCR24 (C and E), and ELOVL2 (C and F) between EC tissues and adjacent normal tissues (n = 187). (G) Scatter plot indicating the correlation between POU2F1 and DHCR24, ELOVL2 based on the IHC results. *P < 0.05, **P < 0.01, ***P < 0.001

The prognostic model based on POU2F1, DHCR24, and ELOVL2 offers a valuable tool for risk outcome prediction in patients with EC

We next assessed the translational value of the prognostic model incorporating POU2F1, DHCR24, and ELOVL2. We first applied the Cox method [20] to calculate the risk score of each patient using the following formula: risk score = POU2F1 expression × 0.832151767023286 + DHCR24 expression × 0.37592797391591 + ELOVL2 expression × 0.497411048330163. Both univariate and multivariate Cox regression analyses confirmed that the risk score served as an independent prognostic factor in patients with EC across the TCGA-UCEC and IHC cohorts (Fig. 8A-B). Stratification on the basis of the risk score clearly distinguished patient outcomes, as patients in the high-risk group had significantly shorter overall survival than those in the low-risk group did (Fig. 8C–D). Importantly, the results of the ROC analyses demonstrated that the model achieved reliable diagnostic efficacy in both cohorts (Fig. 8E–F). These findings highlight the clinical applicability of this three-gene signature, suggesting its potential utility for individualized risk assessment and prognosis-oriented decision-making in EC management (Fig. 8E-F). We also assessed the prognostic value of the model across different histological subtypes and BMI groups. Kaplan‒Meier survival analyses in both the TCGA-UCEC and IHC cohorts demonstrated that the risk score effectively stratified overall survival across multiple clinical subgroups, particularly in the type I EC and BMI-stratified groups (Fig. S5A-B). ROC analyses further confirmed the favorable prognostic performance of the risk score, with consistent predictive accuracy reported across different subgroups (Fig. S5C–D). These findings highlight the clinical applicability of this three-gene signature, suggesting its potential utility for individualized risk assessment and prognosis-oriented decision-making in EC management.

Fig. 8.

The prognostic model constructed based on POU2F1, DHCR24 and ELOVL2 effectively predicts the prognosis of patients with EC. (A-B) Forest plot presenting the results of univariate and multivariate regression analyses based on the risk scores of TCGA-UCEC (A) and IHC (B). (C-D) Survival curves of high- and low-risk scores based on the TCGA-UCEC (C) and IHC (D). (E-F) Diagnostic ROC curves based on TCGA-UCEC (E) and IHC (F)

Discussion

In this study, we identified POU2F1 as a novel transcriptional regulator of cholesterol biosynthesis that promotes EC progression. We demonstrated that POU2F1 directly binds to and activates DHCR24 and ELOVL2, leading to enhanced cholesterol accumulation and malignant phenotypes. These findings establish a mechanistic link between transcriptional regulation and metabolic reprogramming in EC.

Recent studies have shown that metabolic reprogramming is a key hallmark of cancer cells that enables their proliferation, invasion, and metastatic spread [31]. This phenomenon has become a major focus in oncology research. While aerobic glycolysis (the Warburg effect) has traditionally been viewed as the primary metabolic pathway in malignancies, additional data suggest that alternative metabolic mechanisms may also be crucial for certain cancer subtypes [31, 32]. Notably, epidemiological studies have revealed a strong association between metabolic disorders (such as obesity, hyperglycemia, lipid abnormalities, and hypertension) and elevated EC risk in women, underscoring the significant interplay between metabolic dysregulation and endometrial carcinogenesis [33]. Cholesterol homeostasis is crucial for maintaining normal cellular and systemic physiological functions and represents a vital structural and functional component of mammalian cell membranes, where it governs membrane permeability, mechanical stability, and the organization of specialized microdomains [3, 34]. More importantly, cholesterol-rich membrane regions, particularly lipid rafts, serve as critical platforms for signal transduction through the concentration of receptors and signaling molecules [34]. Cellular cholesterol levels are precisely controlled through a sophisticated regulatory system involving biosynthesis, lipoprotein-mediated uptake, reverse transport, enzymatic modification, and intracellular trafficking [3]. Accumulating evidence suggests that cholesterol contributes to oncogenesis through dual mechanisms: by structurally supporting rapid tumor growth and metastasis and by functionally modulating oncogenic signaling pathways that drive proliferation and invasion [3]. In addition to its membrane-related functions, reprogrammed cholesterol metabolism influences multiple cancer hallmarks, including the regulation of autophagic flux, oxidative stress responses, cancer stem cell maintenance, genomic stability, and immunomodulation within the tumor microenvironment [35]. In light of the intricate interplay between cholesterol homeostasis and malignant transformation, elucidating the precise mechanisms of cholesterol metabolic rewiring in EC represents an important research priority.

TFs constitute a specialized class of DNA-binding proteins that recognize and bind to specific promoter or enhancer sequences, orchestrating the initiation of gene transcription through coordinated interactions with RNA polymerase and associated cofactors [36]. These master regulators play indispensable roles in embryogenesis, tissue development, and the maintenance of cellular homeostasis across diverse cell types [37]. Dysregulation of transcriptional control mechanisms frequently contributes to oncogenic transformation and malignant progression [38]. Functionally, TFs may act as either oncogenic drivers or tumor suppressors depending on their cellular context and target genes [38]. Emerging evidence highlights the pivotal involvement of TFs in metabolic reprogramming during endometrial carcinogenesis. For instance, SOX12 has been demonstrated to enhance serine biosynthesis pathways, thereby accelerating EC progression [18]. CENPA promotes glutamine metabolism through transcriptional upregulation of SLC38A1, facilitating tumor growth and aggressiveness [14]. POU2F1, a member of the POU-domain TF family, serves as a critical modulator of diverse biological processes including cellular proliferation, differentiation, and survival [39]. Emerging oncological research has established its role as a potent oncogenic TF across multiple cancer types, including colorectal cancer [11], gastric cancer [40], and cervical cancer [41]. POU2F1 has been shown to promote tumor progression and chemoresistance through metabolic reprogramming [11]. A notable example is its transcriptional upregulation of ALDOA (aldolase A), which enhances glycolytic flux in colorectal cancer [11]. Our integrated bioinformatics and experimental validation revealed that POU2F1 is an important metabolic regulator that orchestrates cholesterol biosynthetic reprogramming in EC. This metabolic reprogramming not only illustrates the mechanism by which POU2F1 enhances malignant progression but also establishes its central role as a key metabolic regulator responsible for the oncogenic rewiring of cellular metabolism.

3β-Hydroxysteroid-Δ24 reductase (DHCR24), the terminal enzyme in the cholesterol biosynthesis pathway, catalyzes the conversion of desmosterol to cholesterol through reduction of the Δ24 double bond [42]. In addition to its canonical role in cholesterol production, DHCR24 participates in diverse cellular processes, including oxidative stress response, cellular differentiation, apoptosis inhibition, and inflammation modulation [43]. Emerging oncological evidence implicates DHCR24 in tumor pathogenesis through its association with enhanced cholesterol biosynthesis and malignant progression across multiple cancer types, notably bladder carcinoma [44], diffuse large B-cell lymphoma [45], and trophoblastic tumors [46]. More importantly, DHCR24 is significantly overexpressed in EC, where it has been mechanistically linked to progesterone resistance—a major therapeutic challenge in this malignancy [5]. Regrettably, the molecular mechanism underlying the overexpression of DHCR24 in EC remains unclear.

The biosynthesis of fatty acids begins with rate-limiting condensation reactions mediated by the elongation of the very-long-chain fatty acid (ELOVL) enzyme family. Current research has characterized seven distinct ELOVL isoforms that exhibit substrate specificity, including saturated/monounsaturated fatty acid specialists (ELOVL1, ELOVL3, ELOVL6, and ELOVL7) and polyunsaturated fatty acid (PUFA)-specific elongases (ELOVL2, ELOVL4, and ELOVL5) [47]. Emerging evidence has demonstrated that elevated ELOVL2 expression significantly contributes to the oncogenic lipid profile and aggressive phenotype in clear cell renal cell carcinoma (ccRCC) [47]. Moreover, ELOVL2 upregulation activates fatty acid metabolic reprogramming to drive enzalutamide resistance in prostate cancer [9]. However, both the functional role and clinical significance of ELOVL2 in EC remain unexplored. Our research revealed that ELOVL2 is highly expressed in EC tissues, as well as in high-grade, late-stage, and the deceased group. Moreover, it is closely associated with poor prognosis. These findings are similar to the clinical significance of ELOVL2 in ccRCC [47], emphasizing the consistent oncogenic role of ELOVL2 in malignant tumors. Our study elucidates the important roles of DHCR24 and ELOVL2 in cholesterol-associated lipid metabolic reprogramming in EC, further highlighting POU2F1 as a key TF in cholesterol biosynthesis reprogramming in this disease. Mechanistic studies indicate that POU2F1 drives cholesterol-associated metabolic reprogramming through the DHCR24/ELOVL2 axis, with DHCR24 mainly contributing to cholesterol biosynthesis and ELOVL2 participating in coordinated lipid remodeling. Unlike DHCR24, which directly catalyzes the terminal step of cholesterol biosynthesis, ELOVL2 does not increase cholesterol production per se. Instead, ELOVL2-mediated elongation of very-long-chain fatty acids facilitates membrane lipid remodeling and lipid droplet biogenesis processes that are essential for efficient cholesterol incorporation, intracellular storage, and cholesterol-dependent oncogenic signaling [48, 49]. Such coordinated regulation of cholesterol biosynthesis and fatty acid elongation represents a common metabolic strategy adopted by rapidly proliferating cancer cells [50, 51]. Notably, the reintroduction of POU2F1 partially restored cholesterol levels and lipid droplet accumulation in DHCR24- or ELOVL2-depleted cells to levels comparable with those in control cells but not to those achieved by POU2F1 overexpression alone. This phenomenon may reflect, at least in part, two non-mutually exclusive mechanisms. First, shRNA-mediated knockdown led to substantial but incomplete depletion of DHCR24 or ELOVL2, resulting in residual expression that may be transcriptionally amplified by POU2F1. Second, as an upstream transcription factor, POU2F1 may coordinate a broader lipid metabolic network rather than acting through a single downstream enzyme, thereby enabling pathway-level compensation via additional POU2F1-responsive genes. Collectively, these findings establish DHCR24 and ELOVL2 as critical yet nonexclusive mediators of POU2F1-driven cholesterol biosynthesis and lipid remodeling, supporting a model in which POU2F1 orchestrates coordinated lipid metabolic reprogramming to sustain malignant progression in EC.

Despite these advances, several limitations should be acknowledged. Although the prognostic value of the POU2F1/DHCR24/ELOVL2 signature was validated in an independent institutional cohort, clinical validation remains limited to a single center with a modest sample size, and larger multicenter studies are warranted. In addition, while our data establish a mechanistic link between POU2F1 and cholesterol-associated lipid metabolic reprogramming, with DHCR24 mediating de novo cholesterol biosynthesis and ELOVL2 contributing to coordinated lipid remodeling, other aspects of cholesterol homeostasis, including uptake, efflux, and intracellular trafficking, were not systematically investigated. Moreover, the upstream signaling pathways regulating POU2F1 expression, its broader downstream transcriptional programs, genome-wide binding landscapes, and potential epigenetic mechanisms that cooperate with POU2F1-mediated transcriptional activation require further exploration. Although in vivo testing of cholesterol-lowering agents such as statins would further enhance the translational relevance of our findings, these pharmacological studies were beyond the scope of the current work and should be addressed in future investigations. Addressing these aspects in future studies will further elucidate the mechanistic and translational significance of POU2F1-driven lipid metabolic reprogramming in EC.

In summary, this study identifies POU2F1 as a central oncogenic regulator in EC through its role in orchestrating coordinated cholesterol biosynthesis and lipid remodeling. Mechanistically, POU2F1 binds directly to and transcriptionally activates DHCR24 and ELOVL2; DHCR24 drives de novo cholesterol biosynthesis, whereas ELOVL2 promotes fatty acid elongation-dependent lipid remodeling. This integrated lipid metabolic program supports cholesterol-dependent signaling, lipid storage, and ultimately the malignant progression of EC. By coupling cholesterol biosynthesis with lipid remodeling, POU2F1 enables metabolic flexibility that sustains tumor aggressiveness, highlighting the POU2F1–DHCR24/ELOVL2 axis as a critical metabolic vulnerability and a potential therapeutic target in EC.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Zi-hui Zhang: Conceptualization, Software, Investigation, Formal analysis, Writing – original draft. Fei-fei Yuan: Data curation, Software, Writing – original draft. Lian Yang: Investigation. Wei Zhang: Visualization. Shuang Li: Investigation, Resources, Writing – review & editing. Yu-qin Huang: Conceptualization, Resources, Supervision, Writing – review & editing.

Funding

This research was sponsored by the Faculty Development Grants from Xiangyang No. 1 People’s Hospital (grant no. 28), the Doctoral Research Start-up Foundation of the Yichang Central People’s Hospital and Natural Science Foundation of Hubei Province (2024AFB962) and the Hunan Provincial Health and Wellness Research Project (20253879).

Data availability

The article contains all original data from this study. Additional inquiries may be addressed to the corresponding authors.

Declarations

Ethics approval and consent to participate

All the experimental animal protocols and the human tissue study were approved by the Ethics Committee of Xiangyang No. 1 People’s Hospital.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.