High glucose promotes aggressive progression of PR-negative endometrial cancer via CD146

Abstract

Background

Hyperglycemia is a risk factor for endometrial cancer (EC) progression, especially in the high-risk progesterone receptor‑negative (PR‑) subtype. A high‑glucose environment can induce epithelial‑mesenchymal transition (EMT) to promote tumor aggressiveness. However, the regulatory mechanism of EMT‑related molecules like CD146 in PR‑ EC under hyperglycemic conditions remains unclear, and its therapeutic potential is undefined. This study aimed to elucidate the role and mechanism of CD146 in this context.

Methods

This study integrated bioinformatics, cellular experiments, and clinical validation. Bioinformatics analysis of the TCGA database assessed CD146 expression in EC, focusing on the PR-low subgroup to evaluate its associations with prognosis, EMT, and glucose metabolism. A high-glucose-induced PR-negative JEC cell model was established for in vitro validation. Gene knockdown and overexpression assays elucidated CD146’s functional role in high-glucose-driven EMT, invasion, and metastasis. Clinical tissue validation was performed using immunohistochemistry (IHC). Finally, drug sensitivity screening combined with molecular docking was employed to identify potential therapeutic compounds targeting CD146.

Results

The results showed that CD146 was markedly downregulated in endometrial cancer (EC) tissues, an effect exacerbated in PR‑negative EC cells under high‑glucose conditions. Low CD146 expression correlated with poor prognosis in PR‑negative EC patients. Mechanistically, CD146 suppression activated epithelial-mesenchymal transition (EMT), accompanied by enhanced cell migration and invasion. Functional studies demonstrated that CD146 knockout accelerated high‑glucose–induced EMT and malignant progression, whereas its overexpression attenuated this process. Immunohistochemical analysis confirmed significantly lower CD146 expression in EC tissues. Furthermore, drug sensitivity screening and molecular docking revealed that CD146‑low EC cells exhibit heightened sensitivity to the TGF‑β pathway inhibitor SB505124, suggesting its pharmacological inhibition as a promising therapeutic strategy for this EC subtype.

Conclusion

In summary, high glucose may promote the progression of PR-negative EC by downregulating CD146 expression and inducing EMT, suggesting that CD146 represents a promising therapeutic target for precision treatment strategies in EC, particularly in patients with concurrent glucose metabolism dysregulation.

Graphical Abstract

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12967-026-08549-5.

Introduction

Endometrial cancer (EC) is one of the most common malignancies of the female reproductive system. Over recent years, both its incidence and mortality have risen steadily, with approximately 69,120 new cases projected in the United States by 2025 [1]. Based on distinct pathogenic mechanisms, EC is classified into two main histopathological subtypes, type I estrogen-dependent and type II estrogen independent [2]. Therapeutic approaches for EC are largely dictated by disease stage. Early-stage disease is typically managed with total hysterectomy, which confers favorable survival outcomes. By contrast, advanced, recurrent, or refractory cases are primarily treated with conventional chemoradiotherapy combined with hormonal therapy. Notably, progesterone receptor (PR)‑negative EC represents a particularly challenging subgroup due to its aggressive tumor biology and intrinsic resistance to progestin-based treatments, leading to markedly inferior prognosis relative to PR‑positive tumors. This therapeutic gap underscores an urgent need for the development of targeted therapeutic strategies specifically designed for PR‑negative EC to improve patient survival and quality of life. Furthermore, as EC management enters the molecular classification era, systematic exploration of novel molecular targets and precision therapeutic strategies has become critically important for advancing personalized treatment approaches.

Accumulating evidence indicates that cancer cells possess an elevated sensitivity to variations in glucose concentration, with their metabolic demands markedly surpassing those of normal somatic cells. Preclinical studies reveal that low-glucose environments suppress cellular proliferation, promote apoptosis, and induce G1-phase cell cycle arrest, whereas hyperglycemic conditions enhance the adhesive and invasive potential of specific malignancies, such as breast and colorectal carcinomas. These observations underscore a strong association between ambient glucose levels and the growth characteristics of these malignancies [3, 4]. In the context of EC, clinical data consistently indicate that hyperglycemia correlates with increased disease incidence and mortality, suggesting that poor glycemic control may serve as a critical factor in EC initiation and progression [5, 6]. Epidemiological analyses further reveal that multiple forms of glucose metabolic dysregulation, including type 2 diabetes, type 1 diabetes, gestational diabetes, and prediabetes, are closely associated with increased EC incidence. Notably, EC patients with concurrent diabetes experience a pronounced elevation in mortality rate, emphasizing the negative prognostic impact of coexisting metabolic disorders on patient outcomes and treatment strategies (Table 1).

Table 1.

Epidemiologic studies linking glucose metabolism disorders and endometrial cancer

Outcome	Glucose Metabolism Disorders / DM	No. of cases	Random Effects (95% CI)	P value	Ref
Incidence	T2DM	16,224	1.65(1.50–1.81)	< 0.05*	[7]
	T2DM	8174	1.97(1.71–2.27)	< 0.001***	[8]
	T2DM	1303	2.60(1.50–4.70)	< 0.05*	[9]
	T1DM	7596	3.15(1.07–9.29)	< 0.05*	[10]
	Prediabetes	966	1.60(1.13–2.27)	0.008**	[11]
	GDM	189	2.69(1.55–4.36)	< 0.001***	[12]
	EDIH	1462	1.58(1.34–1.87)	< 0.001***	[13]
Mortality	DM	26,352	1.42(1.31–1.54)	< 0.05*	[14]
	DM	1644	1.40(1.00-1.80)	< 0.05*	[15]
	T2DM	158	2.73(1.53–4.85)	< 0.05*	[16]
	T2DM	2070	1.32 (1.13–1.55)	< 0.05*	[7]

T2DM: Type 2 Diabetes Mellitus; T1DM: Type 1 Diabetes Mellitus; GDM: Gestational Diabetes Mellitus; DM: Diabetes Mellitus; EDIH: Empirical Dietary Index for Hyperinsulinemia. ^*P<0.05, ^**P<0.01, ^***P<0.001

Epithelial-mesenchymal transition (EMT) is a biological process in which epithelial cells lose polarity and intercellular junctions, undergo cytoskeletal remodeling, and acquire migratory capacity along with a mesenchymal-like transcriptional profile. Studies have established that EMT plays a pivotal role in the initiation and progression of various malignancies [17–19], promotes tumor-stromal interactions, and facilitates invasion and metastasis [20–22]. A high-glucose environment has been recognized as a key driver of EMT in EC. A recent study reported that elevated glucose concentrations suppress the expression of E-cadherin (E-Cad), a key epithelial marker, while concomitantly upregulating Snail, a principal EMT-associated transcription factor, resulting in enhanced cellular adhesion and invasiveness in a subset of PR-positive EC cells [23, 24]. Notably, metformin has been reported to suppress EC proliferation and metastasis [25], and EC patients receiving metformin treatment have demonstrated improved overall survival and progression-free survival outcomes [26, 27]. Collectively, these findings suggest that a high-glucose environment promotes EC malignant progression via induction of EMT, while therapeutic targeting of glucose metabolic pathways may reverse EMT-associated phenotypic alterations and improve clinical outcomes. However, the specific regulatory network linking high glucose to EMT in EC remains poorly defined, and the molecular mediators bridging these processes have yet to be systematically elucidated. In this study, through multidimensional screening and functional validation, we identified CD146 (known as melanoma cell adhesion molecule, MCAM) as a key effector molecule driving EMT progression in EC cells under high-glucose conditions, thereby offering a promising therapeutic target for precision treatment of EC.

This study aimed to elucidate the molecular mechanisms and therapeutic potential of CD146 in PR‑negative EC through an integrated approach encompassing bioinformatics analysis, in vitro cellular functional assays, and clinical histopathological validation. Our findings revealed that CD146 is markedly downregulated in EC tissues, and exposure to a high‑glucose microenvironment further exacerbates this reduction in PR‑negative EC cells. Such suppression of CD146 is closely linked to EMT activation, as evidenced by decreased E‑cadherin and increased N‑cadherin expression, accompanied by a pronounced enhancement in cell migration and invasion. Functional studies demonstrated that CD146 knockout accelerates high‑glucose–induced EMT progression and promotes EC malignant advancement, whereas CD146 overexpression attenuates high‑glucose–mediated EMT activation and suppresses the progression of PR‑negative EC. Building upon these mechanistic insights, by combining drug sensitivity profiling with molecular docking technology, we identified potential small-molecule compounds targeting CD146, thereby providing candidate agents for future therapeutic development. Collectively, these findings suggest that CD146 may serve as a potential therapeutic target for high-risk PR-negative EC associated with glucose metabolic dysregulation. Loss of CD146 expression appears to facilitate tumor progression through EMT induction, thereby offering novel molecular target and intervention strategies for the precision treatment of this refractory subtype of EC from a metabolic perspective.

Materials and methods

Bioinformatics analysis

Differential expression analysis of CD146 in EC

RNA-seq transcriptomic data for EC were retrieved from The Cancer Genome Atlas (TCGA) database. Following rigorous quality control, including data normalization and the removal of duplicate tumor samples, a cohort comprising 545 EC tissues and 35 adjacent non-tumor tissues were established. The expression levels of CD146 (known as MCAM) and PR were extracted to perform differential expression analysis, thereby evaluating their dysregulation in EC. To investigate the combinatorial expression patterns of PR and CD146, the 545 EC tumor samples were stratified into three independent groups based on the tertile distribution of both genes’ expression levels. Using the 33.3% and 66.7% percentiles as cutoffs, samples were categorized into Low, Medium, and High expression subgroups. This study specifically focused on the clinically relevant PR-low subgroup (n = 182), which was further subdivided into PRLow_CD146Low (n = 51), PRLow_CD146Medium (n = 66), and PRLow_CD146High (n = 65). This stratification strategy facilitates a systematic analysis of the synergistic effects between PR and CD146 and allows for an in-depth investigation of the PR-negative patient population. Survival data were sourced from the authoritative study by Liu et al., published in Cell [28], and were utilized for subsequent Kaplan-Meier survival analyses. Furthermore, to validate the observed mRNA expression differences at the protein level, an independent validation was conducted using proteomic data from the Clinical Proteomic Tumor Analysis Consortium (CPTAC).

Investigation of CD146 in glycolysis and EMT in PR-negative EC

Epidemiological studies have established a strong correlation between the development of EC and endocrine metabolic disorders. Notably, individuals with diabetes mellitus face a 1.65-fold increased risk of developing EC compared to the general population [7], suggesting that aberrant glucose metabolism constitutes a pivotal pathogenic factor in EC. Tumor cells sustain their malignant proliferation and metastatic capacity through abnormal nutrient acquisition and metabolic reprogramming, representing a core hallmark of cancer biology. Furthermore, EMT serves as a critical driver of metastasis, playing an indispensable role in the invasive progression of EC. To elucidate the mechanistic link between CD146 and the interconnected processes of metabolic dysregulation and EMT in PR-negative EC, we acquired glycolysis and EMT-related gene sets from the GSEA database (https://www.gsea-msigdb.org/gsea/index.jsp). Pathway activity was quantitatively assessed using a combined algorithm of ssGSEA and PLAGE, implemented via the GSVA package [29]. Through comparative analysis of enrichment scores across distinct CD146 expression cohorts, we preliminarily evaluated the potential regulatory role of CD146 in modulating glycolytic flux and EMT progression in PR-negative EC.

In vitro experimental validation

Bioinformatics analysis revealed that in EC tissues with low PR expression, CD146 may be involved in glycolysis and the EMT process, thereby promoting tumor invasion and metastasis. Moreover, the expression level of CD146 showed a significant impact on patient survival outcomes. To further elucidate the molecular mechanisms of glycolysis and CD146 in PR‑negative EC, the present study conducted systematic in vitro cellular experiments for validation and mechanistic exploration.

Cell culture

The human endometrial carcinoma (EC) cell line JEC was obtained from Crisprbio Biotechnology Co., Ltd. (Beijing, China). Cells were cultured in complete Dulbecco’s Modified Eagle Medium (DMEM) containing 5 mM glucose (Genom BIO), supplemented with 10% fetal bovine serum (FBS, Gibco), 100 U/mL penicillin, and 100 µg/mL streptomycin. All cultures were maintained at 37 °C in a humidified incubator with an atmosphere of 5% CO₂ and 95% air. The cells were passaged twice weekly. The HEC‑1‑A cell line, a PR-negative endometrial carcinoma cell line, was obtained from Wuhan Keep Biotechnologies Co., Ltd. and cultured using a similar protocol. This cell line was specifically selected for subsequent mechanistic investigations into CD146‑induced EMT under high‑glucose conditions.

Wound healing assay

JEC cells were seeded in 6‑well plates and cultured until reaching the required confluence. A straight wound was created in the confluent monolayer by gently scratching with a sterile 200 µL pipette tip, maintaining a scratch width of approximately 0.5–1 mm. The cells were then washed gently 2–3 times with PBS buffer (Solarbio) to remove residual serum components. After aspiration of PBS, serum‑free or low‑serum medium (Biosharp) was added for continued incubation. Images of the wound area were captured under an optical microscope at 0, 24, 48, and 72 h after scratching. The relative cell migration was quantified using ImageJ software.

Cell proliferation assay

Cell viability was assessed using the Cell Counting Kit‑8 (CCK‑8, Bioss) following the manufacturer’s instructions. JEC cells were seeded in 96‑well plates at a density of 3000 cells per well in complete culture medium. After 24 h, the medium was replaced with media containing different glucose concentrations, and the cells were cultured for an additional 24, 48, or 72 h. To measure viability, 10 µL of CCK‑8 solution was added to each well, followed by incubation for 0.5–1 h. The absorbance at 450 nm, reflecting the metabolic conversion of the dye, was measured to determine viable cell density.

Transwell assay

Cell invasion ability was evaluated using Transwell chambers with 8 μm pore size. Briefly, 1.0 × 10⁵ JEC cells were seeded in the upper chamber and allowed to adhere. Cells were then gently washed twice with PBS (Solarbio). The upper chamber was then filled with 100 µL of DMEM medium (Biosharp) containing the following treatments: control (5 mM glucose), high‑glucose 1 (HG1, 10 mM glucose), and high‑glucose 2 (HG2, 25 mM glucose). The lower chamber was supplemented with 600 µL of DMEM medium (Biosharp) containing 10% fetal bovine serum (Zhejiang Tianhang Biotechnology Co., Ltd., China) as a chemoattractant. After 24 h of incubation at 37 °C, cells on the lower surface of the membrane were fixed with methanol (Biosharp) at 37 °C for 20 min and stained with 0.1% crystal violet (BKMAM) for 30 min. The stained invasive cells were observed under an inverted optical microscope at 200× magnification, and three randomly selected fields per chamber were imaged and counted for quantification.

Flow cytometry

Apoptosis in JEC cells was evaluated using flow cytometry. Briefly, 5.0 × 10⁵ JEC cells were seeded into each well of a six‑well plate and cultured until confluent. After attachment, cells were gently washed two to three times with phosphate‑buffered saline (PBS; Solarbio). Cells were then treated with DMEM medium supplemented with different experimental conditions: control group, HG1 group, and HG2 group. Cultures were maintained at 37 °C in a humidified atmosphere containing 5% CO₂ for 24, 48, or 72 h. At the end of each treatment period, culture supernatants were harvested. Attached cells were detached by incubation with EDTA‑free trypsin (Solarbio) and combined with their corresponding supernatants. Cell suspensions were pelleted by centrifugation at 1000 rpm for 5 min at room temperature, washed twice with ice‑cold PBS under the same centrifugation conditions, and then resuspended in pre‑chilled 1× Binding Buffer (Solarbio) to achieve a cell density of 1 × 10⁶–5 × 10⁶ cells/mL. For apoptosis staining, 100 µL of the cell suspension was incubated with 5 µL Annexin V‑FITC (Solarbio) and 5 µL propidium iodide (PI; Solarbio) at room temperature in the dark for 5 min. Following incubation, 400 µL of pre‑chilled 1× Binding Buffer (Solarbio) was added, and samples were immediately analyzed by flow cytometry within 1 h.

Western blot analysis

Total cellular proteins were harvested from JEC cells using a radioimmunoprecipitation assay (RIPA) lysis buffer (Solarbio, China) on ice. The concentration of extracted proteins was measured using a BCA protein assay kit (Solarbio, China). Samples containing 20–50 µg of total protein were denatured and subjected to separation by 10% SDS-PAGE, followed by wet transfer onto PVDF membranes (Solarbio, China). To minimize non-specific binding, the membranes were blocked with 5% non-fat dry milk dissolved in TBST for 2 h at room temperature. Subsequently, the membranes were incubated with specific primary antibodies: anti-N-Cadherin (1:5000), anti-E-Cadherin (1:20,000), and anti-CD146 (1:5000) (Proteintech, China), at 4 °C overnight with gentle agitation. After extensive washing, the membranes were incubated with HRP-conjugated secondary antibodies for 1 h at room temperature. Immunoreactive bands were detected using a Bio-Rad Gel Doc XR system and densitometric analysis was performed using ImageJ software (Bethesda, MD, USA).

Quantitative real-time PCR

Total RNA was extracted from cultured cells using TRIzol reagent (Invitrogen) following the manufacturer’s protocols. RNA concentration and purity were determined spectrophotometrically using a NanoPhotometer (Thermo Scientific, MA, USA). Reverse transcription reactions for mRNAs were carried out using FastKing RT Kit (Tiangen, China). Quantitative real-time PCR amplification was performed using SYBR Green chemistry (Takara, China) on an Applied Biosystems 7500 Real-Time PCR System (Life Technologies). Specific primers used for target genes are detailed in Supplementary Table 1. Relative quantification of gene expression was performed using the 2 − ΔΔCt method, normalized to the reference gene GAPDH.

CD146 Gene knockdown and overexpression experiments

To manipulate CD146 (MCAM) expression in JEC cells, cells were transfected with either human CD146(MCAM)-specific siRNA or pcDNA3.1-CD146༈MCAM༉-3xFlag plasmid (Abiowell). Cells were cultured to appropriate confluency and density, ensuring they were in exponential growth phase with healthy morphology prior to transfection. Approximately 5–6 h before transfection, culture medium was replaced with serum‑free medium (Biosharp) to deprive serum components that may interfere with complex formation; at the time of transfection, cells reached 70%–90% confluence and were washed twice with phosphate‑buffered saline (PBS). Transfections were performed in six‑well plates. For each well, solution A was prepared by diluting Lipofectamine 3000 (Solarbio) in Opti‑MEM (BKMAM), and solution B by diluting siRNA (HonorGene) or plasmid DNA (Abiowell) in the same medium. After gentle mixing, both solutions were incubated separately for 5 min at room temperature, then combined and further incubated for 20 min to allow formation of transfection complexes. The mixture was added dropwise onto the cell monolayer, and 6–8 h later, the medium was replaced with complete growth medium. Cells were harvested 48 h post‑transfection for analysis. Knockdown and overexpression efficiencies were assessed by qRT‑PCR and Western blot analysis, and subsequent gene and protein expression analyses were conducted accordingly.

Immunohistochemical validation of differential expression

To validate the differential expression of CD146 in EC, pathological specimens were retrospectively collected from the Department of Pathology at Jingzhou First People’s Hospital between 2022 and 2025. The cohort included 80 EC tumor tissues and 40 adjacent non-tumor tissues, with 20 paired samples available for analysis. Immunohistochemical (IHC) staining was performed to quantitatively assess CD146 expression levels. The protocol strictly adhered to the manufacturer’s instructions using the IHCeasy CD146/MCAM Ready-to-Use IHC Kit (Proteintech, Cat No: KHC2274). Stained slides were visualized under an OLYMPUS BX53 microscope. Semi-quantitative scoring was conducted by a board-certified pathologist utilizing ImageJ software integrated with the IHC Profiler plugin. The final score was calculated by multiplying two parameters: staining intensity (0 = no staining; 1 = weak; 2 = moderate; 3 = strong) and the percentage of positive cells (0 = 0–5%; 1 = 5–25%; 2 = 26–50%; 3 = 51–75%; 4 = 76–100%). Statistical analysis was subsequently performed on these scores to evaluate the differential expression of CD146 between cancerous and para-cancerous tissues [30–32]. This study was approved by the Medical Ethics Committee of Jingzhou First People’s Hospital.

CD146 Drug sensitivity analysis and molecular docking

To elucidate the pharmacological relevance of the CD146 locus, we integrated drug sensitivity analysis with structure-based molecular docking. Initially, drug sensitivity scores were computed using the oncoPredict R package in conjunction with the Genomics of Drug Sensitivity in Cancer 2 (GDSC2) database to identify the top five candidate compounds exhibiting the highest correlation [33, 34]. Subsequently, the three-dimensional structures of these small molecules were retrieved from the PubChem database, while the CD146 protein structure was predicted with high fidelity using the AlphaFold algorithm [35, 36]. Molecular docking was then performed using AutoDock Vina 1.2.7, adhering to its standard protocol with supplementary parameter optimization to ensure accuracy. The resulting protein-ligand complexes were subjected to detailed interaction analysis using the Protein-Ligand Interaction Profiler (PLIP) to characterize critical intermolecular forces [37–39]. Finally, the binding conformations and interaction patterns were visualized in three dimensions using the PyMOL Molecular Graphics System (Version 3.1.0, open-source PyMOL).

Statistical analysis

Statistical analyses were performed using SPSS software (version 26.0). Quantitative data are expressed as the mean±standard deviation (SD). Differences in means between groups were compared using Student’s t-test or the Mann-Whitney U test for non-parametric data, and one-way analysis of variance (ANOVA) for comparisons involving more than two groups. Data visualization was conducted using GraphPad Prism software (version 9.0). Western blot band intensities and wound healing areas were quantified using ImageJ software [32]. Each experiment was repeated independently at least three times. Bioinformatic analyses were performed using the R programming language (version 4.5.0). A P-value of < 0.05 was considered statistically significant.

Results

Investigation of CD146 in Glycolysis and EMT in EC

To investigate the differential expression of CD146 in EC, we analyzed RNA-seq data from 545 EC tissues and 35 adjacent normal tissues obtained from TCGA database. Compared with normal endometrium, both PR and CD146 mRNA levels were significantly downregulated in EC tissues (Fig. 1A, B). This downregulation was further validated at the protein level using proteomic data from the CPTAC, which confirmed markedly reduced expression of both PR and CD146 in EC compared with normal endometrial samples (Supplementary Fig. 1), supporting a concordant mRNA-protein expression pattern. Given that both PR and CD146 exhibit a co-downregulation pattern in EC, we further investigated their potential clinical significance in tumor progression. Based on the expression levels of PR and CD146, EC cases were systematically classified into subgroups (Fig. 1C). In the PR-low expression subgroup, Kaplan-Meier survival analysis was performed to assess the impact of CD146 expression status on patient clinical outcomes. The results revealed that CD146 expression level showed no significant correlation with overall survival (OS) in this subgroup (p = 0.1785, Fig. 1D), but was significantly associated with disease-specific survival (DSS) and progression-free interval (PFI) (p < 0.05, Fig. 1E, F). Multiple-testing analysis further demonstrated that both high and low CD146 expression levels were associated with DSS and PFI, with low CD146 expression exhibiting a particularly strong effect (Fig. 1G). Next, to explore the potential biological pathways through which CD146 affects patient prognosis, we conducted pathway enrichment analysis. Pathway enrichment analysis indicated that, in PR-low EC patients, CD146 expression was closely linked to EMT and glucose metabolism-related pathways (Fig. 1H-K). To verify whether these pathways mediate the biological functions of CD146, we further explored the underlying mechanisms through functional cellular experiments.

Fig. 1.

CD146 is downregulated in EC, and its prognostic value and regulatory roles in glucose metabolism and EMT. A: Expression of CD146 (MCAM) in EC tissues. B: Expression of progesterone receptor (PR) in EC tissues. C: Systematic stratification of EC samples based on expression levels of PR and CD146. D-F: Survival analyses for overall survival (OS), disease‑specific survival (DSS), and progression‑free interval (PFI). G: Results of post‑hoc multiple comparisons. H-J: Relationship between CD146 expression and the EMT process in EC tissues. K: Effect of CD146 on the GLYCOLYSIS_FRU26BP_REGULATION metabolic process. *P<0.05, **P<0.01, ***P<0.001

High glucose promotes proliferation, invasion, and migration of JEC cells, while inhibiting apoptosis

The above results indicate that CD146 is downregulated in EC tissues and is involved in modulating EMT and glucose metabolism. Nevertheless, the precise regulatory interplay and molecular mechanisms linking CD146, EMT, and glucose metabolism remain to be elucidated. To address this, we carried out a series of cell-based functional experiments.

To investigate the influence of a high-glucose environment on JEC cells, we first evaluated proliferative activity under glucose stimulation. The CCK-8 assay revealed that treatment with 10, 25, and 50 mM glucose significantly increased the proliferation of JEC cells in a concentration-dependent manner compared to the control (Fig. 2A), indicating that elevated glucose levels substantially enhance cell proliferation. Wound-healing assays showed that exposure to 25 mM glucose markedly elevated the migration rate of JEC cells (Fig. 2B, C), suggesting enhanced migratory capacity. Similarly, Transwell invasion assays demonstrated that 25 mM glucose treatment robustly increased the invasive ability of JEC cells (Fig. 2D, E), further supporting a pro-invasive role of high glucose. Furthermore, flow cytometric analysis revealed that treatment with both 10 mM and 25 mM glucose significantly reduced the apoptotic rate in JEC cells (Fig. 2F, G), implying that high glucose also exerts an anti-apoptotic effect.

Fig. 2.

High glucose promotes proliferation and invasion, and inhibits apoptosis in JEC cells. A: Comparison of JEC cell proliferation after treatment with different concentrations of high glucose for 24 h, 48 h, and 72 h. B-C: Comparison of invasive ability of JEC cells treated with different concentrations of high glucose. D-E: Comparison of migration ability of JEC cells treated with different concentrations of high glucose. F-G: Comparison of apoptosis in JEC cells treated with different concentrations of high glucose, as measured by flow cytometry. HG1: 10mM, HG2: 25mM, HG3: 50mM. *P<0.05 vs. Control group, #P<0.05 vs. HG1 group

In summary, these results demonstrate that a high-glucose microenvironment inhibits apoptosis and enhances the proliferative, migratory, and invasive capacities of JEC cells, suggesting that elevated glucose levels may contribute to EC progression by fostering a more aggressive tumor phenotype.

High glucose downregulates CD146 expression and promotes EMT in JEC cells

To further investigate the effects of a high glucose on CD146 expression and EMT in JEC cells, we established a high-glucose-induced cell model and examined changes in key EMT markers. First, we examined whether high glucose effectively induces the EMT process in JEC cells. qRT-PCR results revealed that high glucose treatment significantly upregulated the expression of mesenchymal associated markers, including N cadherin, Vimentin, Snail, and Slug, while downregulating the epithelial marker E cadherin at the mRNA level(Fig. 3A). Consistently, CD146 expression was also markedly decreased under high‑glucose conditions (Fig. 3B). Western blot analysis further confirmed that high‑glucose exposure not only significantly upregulated N‑cadherin protein levels, a defining hallmark of EMT, but also coincided with a pronounced reduction in CD146 expression (Fig. 3C, D).These results indicate that CD146 expression is negatively correlated with the high‑glucose‑induced EMT, suggesting that CD146 may serve as a regulatory factor involved in this process.

Fig. 3.

High glucose inhibits CD146 expression and promotes EMT in JEC cells. A-B: Expression levels of CD146 and the mesenchymal markers (N-cadherin, Vimentin, Snail, and Slug) and epithelial marker (E-cadherin) in JEC cells cultured with different concentrations of high glucose, as determined by qRT-PCR. C-D: Protein expression levels of CD146 (C) and E-cadherin (D) in JEC cells cultured with different concentrations of high glucose were determined by Western blot. HG1: 10 mM, HG2: 25mM. ^*P<0.05 vs. Control group

High glucose facilitates EMT in JEC cells by downregulating CD146

Further Western blotting demonstrated that CD146 knockdown potentiated the high glucose-induced upregulation of N-cadherin and led to a corresponding suppression of E-cadherin, supporting a role for CD146 depletion in promoting EMT and accelerating tumor progression under high glucose conditions (Fig. 4A–D). In contrast, CD146 overexpression markedly lowered N-cadherin and elevated E-cadherin levels, indicating that enforced CD146 expression can effectively suppress high glucose-induced EMT in JEC cells (Fig. 4E–H). Consistent with the observations in JEC cells, CD146 knockdown in HEC-1 A cells yielded analogous results (Supplementary Fig. 2A-C). Specifically, while CD146 depletion further potentiated the high-glucose-induced upregulation of N-cadherin, it concurrently suppressed E-cadherin expression under normal glucose conditions. Furthermore, a marked morphological transition was observed (Supplementary Fig. 2D), characterized by a shift from an epithelial cobblestone or polygonal morphology to a mesenchymal spindle-shaped phenotype, a characteristic morphological feature of EMT process.

Fig. 4.

Validation of the effect of CD146 on EMT in JEC cells under high-glucose conditions. A: Western blot analysis of N‑cad, E‑cad, and CD146 protein expression following CD146 knockdown. B-D: Quantitative statistical analysis of CD146, N‑cad, and E‑cad protein levels under different treatment conditions. HG:25mM. *P<0.05 vs. NC group, #P < 0.05 vs. NC + HG group. E: Western blot analysis of N‑cad, E‑cad, and CD146 protein expression following CD146 overexpression. F-H: Statistical analysis of CD146, N‑cad, and E‑cad protein levels under different treatment conditions. HG:25mM. *P<0.05 vs. NC group. I: Representative immunohistochemical (IHC) staining images of CD146 in tumor tissues. Top: 100 μm, bottom: 50 μm. J: Representative IHC images of CD146 in normal tissues. Top: 100 μm, bottom: 50 μm. K: Overall differential expression of CD146 in pathological tissues. L: Differential expression of CD146 in paired samples. ^*P<0.05. ^**P<0.01

Collectively, these findings indicate that loss of CD146 expression under high glucose conditions may exacerbate EMT in EC cells, thereby promoting tumor proliferation and migration. High glucose stimulation may facilitate EMT and enhance the malignant phenotype of EC cells by regulating the expression of the cell adhesion molecule CD146.

Clinical validation of CD146

To further validate the expression and role of CD146 in EC, we conducted IHC analysis. The results showed that CD146 expression was significantly lower in EC tissues compared with normal endometrial tissues. Semiquantitative scoring and statistical analysis confirmed that CD146 levels were markedly reduced in EC specimens relative to normal controls (P < 0.01, Fig. 4I–K). Moreover, paired‑sample analysis further demonstrated consistently lower CD146 expression in EC tissues than in matched adjacent normal tissues from the same patients (P < 0.01, Fig. 4L). To this point, multi‑level analyses spanning transcriptomic, proteomic, and tissue sample investigations have consistently confirmed the down‑regulation of CD146 expression in EC. These clinical observations are consistent with the preceding in vitro findings, reinforcing the downregulation of CD146 in EC.

CD146 Drug sensitivity analysis and molecular docking

To evaluate the potential of CD146 as a therapeutic target in EC, we integrated drug sensitivity screening with molecular docking approaches to systematically explore its pharmacological role. Building on the finding that low expression of CD146 is associated with poor prognosis in EC patients, we further explored therapeutic strategies targeting CD146‑related signaling pathways through drug‑sensitivity analysis. Based on the Spearman’s correlation (ρ) and Cliff’s delta (δ) between gene expression and drug sensitivity, we identified five small-molecule compounds that were most significantly associated with CD146 expression. These included the Mcl‑1 inhibitor UMI‑77_1939, the Aurora kinase inhibitor Tozasertib_1096, the TGF‑β receptor inhibitor SB505124_1194, the VEGFR inhibitor Cediranib_1922, and the FGFR inhibitor PD173074_1049, which are involved in key biological processes such as apoptosis, cell cycle regulation, EMT, angiogenesis, and growth factor signaling (Fig. 5A).

Fig. 5.

Drug sensitivity analysis and molecular docking based on CD146. A: Volcano plot summarizing overall drug sensitivity for screening candidate compounds. B-F: Scatter plots showing correlation between CD146 expression and predicted sensitivity (IC₅₀) to individual drugs, in the order of PD173074, Cediranib, UMI‑77, Tozasertib, and SB505124. G-K: Molecular docking models of CD146 with the five small‑molecule compounds listed above, in the same order. The insets in the upper‑left corner illustrate detailed interaction types and ligand molecular structures

Further analysis in the PR-low subgroup demonstrated a significant association between CD146 expression and predicted drug sensitivity (FDR < 0.001). Specifically, CD146 levels were negatively correlated with sensitivity to PD173074_1049 (ρ = −0.50, δ = −0.52, Fig. 5B), Cediranib_1922 (ρ = −0.49, δ = −0.46, Fig. 5C), UMI-77_1939 (ρ = −0.45, δ = −0.46, Fig. 5D), and Tozasertib_1096 (ρ = −0.43, δ = −0.42, Fig. 5E), suggesting that higher CD146 expression may enhance tumor sensitivity to these agents. Notably, CD146 expression exhibited a significant positive correlation with sensitivity to the TGF-β receptor inhibitor SB505124_1194 (ρ = 0.42, δ = 0.41, Fig. 5F), indicating that lower CD146 expression may be associated with enhanced sensitivity to this compound. These computational findings align with our previous in vitro observations that CD146 downregulation promotes EMT in JEC cells under high-glucose conditions. Collectively, these data suggest that low CD146 expression may foster a cell state dependent on TGF-β signaling to sustain an invasive phenotype. Consequently, targeting this pathway with SB505124 may represent a potential therapeutic strategy for this specific molecular subtype of EC.

Molecular docking analysis further supported the above findings, demonstrating that all evaluated small-molecule compounds bound to the CD146 protein with favorable affinity, as evidenced by binding energies below − 5.00 kcal·mol⁻¹ and the formation of at least two hydrogen bonds (Table 2; Fig. 5G‑K). These results indicate favorable binding affinity and stable interaction patterns between the compounds and CD146, thereby providing a structural foundation for the future design and optimization of CD146-targeted therapeutics.

Table 2.

From drug sensitivity to mechanism: molecular docking of active CD146 ligands

Target Protein	Ligand	Hydrogen-Bond Forming Residues	Binding Free Energy (ΔG, kcal·mol⁻¹)
CD146	PD173074	GLN-171 GLN-171	-6.524
CD146	Cediranib	ILE-427 GLN- 478	-6.812
CD146	UMI.77	GLY-403 ARG-479 GLN-478	-6.704
CD146	Tozasertib	TYR-175 GLU-224	-7.395
CD146	SB505124	LYS-176 ASN-177 GLU-184	-6.918

Discussion

EC represents one of the most prevalent gynecological malignancies worldwide. Over the past decade, the global incidence and mortality rates of EC have shown a significant upward trend. The current standard of care for EC remains primarily surgical, often supplemented with adjuvant radiotherapy and chemotherapy in a multimodal approach. This treatment modality has achieved relatively favorable outcomes in patients with early-to-mid-stage EC and in those without fertility preservation needs [40]. However, therapeutic efficacy remains unsatisfactory for patients with fertility preservation requirements or those diagnosed with advanced-stage disease, particularly among PR-negative EC patients. Therefore, there is an urgent need to explore more diverse and precisely targeted therapeutic strategies to provide novel perspectives and solutions for the individualized treatment of EC. Clinical data indicate that hyperglycemia is associated with increased EC incidence and mortality, suggesting that poor glycemic control may constitute a critical driver of EC pathogenesis and progression. It is noteworthy that hyperglycemia often coexists with intestinal microbiota dysbiosis, exhibiting a bidirectional regulatory relationship: hyperglycemia can induce intestinal microbiota disorders, while the dysregulated microbiota can further exacerbate blood glucose abnormalities by affecting host metabolic and immune homeostasis. This vicious cycle collectively promotes systemic metabolic dysfunction and even provides a favorable microenvironment for tumorigenesis and progression, thereby accelerating tumor development [41, 42]. Moreover, a high-glucose environment can induce EMT, thereby promoting tumor proliferation, invasion, and metastasis [43–45]. Nevertheless, the specific regulatory network linking high glucose to EMT in EC remains poorly defined, and the molecular mediators bridging these processes are yet to be systematically elucidated. Within this framework, the present study, employing multidimensional screening combined with functional validation, has identified CD146 as a key effector molecule driving EMT progression in PR-negative EC cells under high-glucose conditions. More importantly, this study reveals that a high‑glucose environment drives EMT by targeting CD146, thereby establishing a direct molecular link between the clinically common hyperglycemia/diabetes state and endometrial cancer aggressiveness. These findings not only provide a mechanistic explanation for the poorer prognosis observed in endometrial cancer patients with diabetes, but also highlight the value of CD146 as a potential prognostic biomarker in metabolism‑dysregulated tumors. From a therapeutic perspective, this mechanism suggests that interventions targeting CD146 and its downstream signaling pathways may offer novel treatment strategies for this subtype of metabolism‑associated aggressive endometrial cancer, laying an important theoretical foundation for its translation into a clinical therapeutic target.

Our findings revealed that CD146 is markedly downregulated in EC tissues, and this reduction is further exacerbated in PR-negative EC cells under a high-glucose microenvironment. Such suppression of CD146 is closely linked to EMT activation, as evidenced by decreased E‑cad expression alongside increased levels of N-cadherin, Vimentin, and the transcription factors Snail and Slug. This phenotypic shift is accompanied by a marked enhancement in cell migratory and invasive capabilities. Functional studies demonstrated that CD146 knockout accelerates high‑glucose–induced EMT and promotes malignant progression in EC, whereas CD146 overexpression attenuates high‑glucose–mediated EMT activation and suppresses progression in PR‑negative EC. These results collectively suggest that the pro‑EMT effects of high glucose are mediated through CD146 downregulation. In terms of disease relevance, the mechanism elucidates the biological link between a hyperglycemic state (particularly observed in PR‑negative subtypes) and poor prognosis in endometrial cancer patients. Furthermore, it provides a direct theoretical basis for reversing high‑glucose‑driven tumor progression by targeting CD146 and its downstream signaling pathways.

To further validate the expression pattern of CD146 in clinical EC specimens, we analyzed postoperative pathological tissue samples from EC patients. Immunohistochemical and molecular analyses consistently confirmed a significant decrease in CD146 expression in EC tissues compared to adjacent non‑tumor tissues, reinforcing its potential as a therapeutic target. Building on these mechanistic insights, we integrated drug sensitivity profiling with molecular docking approaches and identified potential candidate small‑molecule compounds that interact with CD146, thereby offering candidate agents for future therapeutic development. The results showed that low CD146 expression may foster a cell state dependent on TGF-β signaling to sustain an invasive phenotype. Consequently, pharmacological inhibition of this pathway using SB505124 represents a promising therapeutic strategy for this specific molecular subtype of EC. For PR‑negative EC patients exhibiting low CD146 expression, particularly those refractory to conventional therapies, TGF‑β pathway inhibitors such as SB505124, may represent a promising precision treatment option. Taken together, our study identifies CD146 as a key regulator of glucose‑mediated EMT in PR‑negative EC and underscores its translational potential as a therapeutic target in high‑risk EC associated with dysregulated glucose metabolism.

CD146 is a type I transmembrane glycoprotein and Ca²⁺-independent cell adhesion molecule belonging to the immunoglobulin superfamily. It is closely associated with diverse pathological conditions, including tumorigenesis, inflammatory responses, and autoimmune diseases. Functioning not only as a tumor marker but also as a key membrane receptor, CD146 modulates multiple signaling pathways, thereby influencing critical cellular processes such as tumor cell proliferation, EMT, invasion, and migration [46, 47].

During tumor initiation and progression, CD146 expression demonstrates dynamic alterations. Although the expression patterns of CD146 and its influence on tumor biology vary across different cancer types, and its clinical translational potential remains incompletely defined, accumulating evidence supports a critical regulatory role for CD146 in the EMT process [48]. For instance, in breast cancer, CD146 is highly expressed, and its upregulation activates YAP and modulates the LATS1 pathway, thereby enhancing tumor stemness and promoting EMT [49]. Similarly, in hepatocellular carcinoma, CD146 is significantly upregulated, where it accelerates EMT by promoting IL‑8 expression and suppressing STAT1 [50]. Interestingly, contrasting with these findings, CD146 expression is downregulated in colorectal cancer (CRC), and its overexpression notably inhibits N‑cadherin and MMP2/9 expression, suggesting a tumor‑suppressive role for CD146 in CRC potentially through EMT suppression [51]. Further mechanistic investigations in CRC have revealed that CD146 knockdown suppresses NF‑κB/p65‑induced GSK‑3β expression, thereby promoting β‑catenin nuclear translocation and activation, enhancing tumor stemness, and accelerating disease progression [52, 53]. It is noteworthy that existing literature generally reports that CD146 is highly expressed in EC and plays a cancer‑promoting role. However, these studies are mostly conducted under normal metabolic conditions or focused on specific cell types (e.g., CD146 + cancer‑associated fibroblasts), suggesting that CD146 may have distinct functions in normometabolic contexts or within particular cellular subsets [54, 55]. Notably, gene function is inherently complex and context‑dependent, with expression profiles and biological roles subject to significant modulation by distinct metabolic states. The observed variations in CD146 expression across different malignancies reflect a close relationship between its function and the tumor microenvironment, particularly local glucose metabolic dynamics. Consequently, this study specifically focuses on the regulatory mechanisms by which a high-glucose microenvironment influences the expression and function of CD146, aiming to elucidate its biological role under conditions of metabolic duress. In the present study, we observed a low expression trend of CD146 in EC, which aligns with the expression pattern and certain functional characteristics previously reported in CRC. This suggests that CD146 may similarly function as a tumor suppressor in EC. Further analysis indicates that CD146 downregulation may influence EMT‑related and glucose metabolism‑associated signaling pathways, thereby modulating the malignant behavior of EC.

These findings not only enhance our understanding of the tissue‑specific functions of CD146 in oncology but also provide novel experimental evidence for elucidating the pathogenesis of EC. At the clinical level, CD146 could be established as a glucose‑responsive regulatory factor, bridging the knowledge gap between metabolic dysregulation and EC progression, thereby offering a mechanistic basis for the poor prognosis observed in EC patients with comorbid diabetes. Furthermore, the differential drug sensitivity associated with CD146 expression levels, notably the augmented responsiveness of CD146‑low tumors to TGF‑β pathway inhibitors, highlights a viable strategy for patient stratification and personalized therapeutic intervention. This suggests that CD146 expression status may serve as a predictive biomarker to guide treatment selection, especially for PR‑negative EC patients with concurrent hyperglycemia who often exhibit limited responses to conventional therapies. Consequently, these results lay a crucial theoretical foundation for developing therapeutic strategies targeting CD146 and its associated signaling networks, thereby facilitating the clinical translation of precision interventions in high‑risk EC cohorts.

This study has certain limitations. In terms of mechanistic exploration, we systematically analyzed the potential biological functions of CD146 in EC using bioinformatics approaches and confirmed in vitro that a high-glucose environment downregulates CD146 expression, thereby promoting EMT and enhancing the invasive and migratory capabilities of PR‑negative EC cells. However, these findings necessitate further in vivo validation using animal models. With respect to therapeutic investigation, while we computationally identified potential CD146-targeting agents, rigorous empirical validation of their bioactivity remains to be conducted. Therefore, future research should prioritize elucidating the regulatory mechanisms governing CD146 in vivo through well-designed animal studies and systematically validating candidate compounds identified from computational screening. Such endeavors will provide more robust preclinical evidence for developing CD146-directed therapeutic strategies for EC, potentially expanding treatment options and offering novel insights into the precise prevention and management of EC from a metabolic perspective.

Conclusion

This study employed an integrated research strategy combining bioinformatics analysis with in vitro experimental validation to systematically investigate the biological functions and clinical significance of CD146 in EC. Our findings indicate that low CD146 expression is significantly correlated with poor prognosis in PR-negative EC patients, potentially impacting disease progression by modulating signaling pathways associated with EMT and glucose metabolism. Further mechanistic studies demonstrated that a high-glucose microenvironment downregulates CD146 expression in JEC cells, which in turn drives EMT and enhances the invasive and metastatic capacity of tumor cells. Drug sensitivity screening coupled with molecular docking analyses revealed that EC cells with low CD146 expression exhibit markedly increased sensitivity to the TGF-β signaling pathway inhibitor SB505124, implying that targeting the TGF-β pathway may effectively counteract the EMT state induced by CD146 downregulation. In summary, these findings position CD146 as a promising therapeutic target for precision treatment strategies in EC, especially in patients with coexisting glucose metabolism dysregulation, thereby providing a potential direction for personalized and clinically translatable interventions.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

EC

Endometrial cancer

EMT

Epithelial-Mesenchymal Transition

PR

Progesterone Receptor

CD146(MCAM)^★

Cluster of Differentiation 146(Melanoma Cell Adhesion Molecule)★In this study, for clarity and consistency, CD146​ is used uniformly throughout the text. In analyses involving original data from sources such as TCGA, the corresponding gene symbol MCAM​ will be noted

Author contributions

WL: Bioinformatics analysis, Writing – original draft. WM: Investigation (Cell experiments), Data curation, Validation, Writing – original draft. QZ: Conceptualization, Supervision, Writing – review & editing. XZ: Investigation (Cell experiments), Data curation, Validation, Writing – original draft. DK: Investigation (Pathology experiments), Validation, Writing – review & editing. CY: Supervision. JT: Conceptualization, Supervision, Software, Writing – review & editing. YW: Conceptualization, Resources, Supervision, Writing – review & editing. All authors read and approved the final manuscript.

Funding

This work was supported by grants from the Joint Scientific Research Fund of Jingzhou (Grant No. 2024LHY26), the Hubei Provincial Health Science and Technology Project (Grant No. WJ2025M098), and the Natural Science Foundation of Hubei Province (Grant No. 2026AFC0558).

Data availability

The original contributions presented in this study are included in the article and/or the Supplementary Material. Further inquiries can be directed to the corresponding author(s).

Declarations

Ethics approval and consent to participate

This study was reviewed and approved by the Medical Ethics Committee of Jingzhou First People’s Hospital (No. KY202426). The ethics committee waived the requirement for written informed consent for the use of human pathological tissues, in accordance with local legislation and institutional guidelines. All experiments involving JEC and HEC-1-A cell lines were performed in accordance with standard biosafety and laboratory protocols.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.