Single-Cell analysis Uncovers the Dynamic Process of EMT in Endometrial Cancer and the Core Roles of INHBA and POSTN in Regulating Tumor Malignant Phenotypes

Single-Cell analysis Uncovers the Dynamic Process of EMT in Endometrial Cancer and the Core Roles of INHBA and POSTN in Regulating Tumor Malignant Phenotypes | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Single-Cell analysis Uncovers the Dynamic Process of EMT in Endometrial Cancer and the Core Roles of INHBA and POSTN in Regulating Tumor Malignant Phenotypes Dan Liu, Ben Wang, Jie Huang, Ge Diao, Yunhe Ma, Xuena Chen, Runbo Li, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8267986/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Endometrial cancer (EC) incidence rises with aging and obesity epidemics. While early-stage EC has favorable outcomes, metastatic disease portends poor survival, necessitating biomarkers for metastasis prediction. Epithelial-mesenchymal transition (EMT) drives EC progression, but its dynamics and core regulators in clinical specimens remain uncharacterized.We performed single-cell RNA sequencing on 7 EC, integrated with bulk RNA-seq from 521 TCGA-UCEC cases. Cellular trajectories were reconstructed via pseudotime analysis. A specific 47 gene signature was derived by consensus non-negative matrix factorization. Core EMT regulators were functionally validated in vitro. We mapped a sequential EMT progression axis in clinical specimens, identifying a transitional hybrid E/M state. A specific 47-gene signature stratified patients into two groups with significantly divergent survival; functional validation pinpointed INHBA and POSTN as core EMT drivers, whose knockdown suppressed migration and invasion, reversed EMT, and impaired proliferation, with stage-dependent upregulation confirmed in advanced EC. This study defines INHBA and POSTN as central EMT regulators with prognostic and therapeutic potential, providing the first map of EMT dynamics in clinical EC specimens. Endometrial cancer single cell sequencing epithelial-mesenchymal transition INHBA POSTN Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Endometrial cancer (EC) has emerged as a leading malignancy of the female reproductive tract and represents an escalating global health burden. As highlighted in recent reviews, the rising incidence of EC is closely linked to two major drivers: the global aging population and the obesity epidemic, which have driven a steady increase in new diagnoses over the past decade [1,2]. Clinically, approximately two thirds of patients are diagnosed at an early-stage, and exhibit a favorable prognoses [1]. However, the prognosis diverges sharply for patients with metastatic or recurrent EC: despite advances in systemic treatments, including chemotherapy and targeted therapy, survival rates remain below 20%, highlighting the limitations of current treatment strategies [2]. This critical gap between early-stage success and advanced disease failure has prompted an urgent demand to identify molecular biomarkers that can predict risk of metastasis, stratify prognosis, and guide targeted interventions—efforts that are central to addressing the unmet needs of EC patients. Epithelial–mesenchymal transition (EMT) is a biological process in which epithelial cells acquire mesenchymal phenotypes through a series of coordinated steps. During EMT, epithelial cells progressively lose defining features such as E-cadherin and keratin expression, while gaining mesenchymal markers including N-cadherin and vimentin. EMT plays diverse roles in embryonic development, wound healing, tissue regeneration, and tumor progression. In cancer, it is increasingly recognised that EMT underpins metastasis, stemness, and drug resistance[3]. Rather than a binary shift, EMT is now understood as a dynamic continuum, during which hybrid E/M cells emerge that exhibit both epithelial and mesenchymal traits, thereby facilitating dissemination and colonisation[4]. Whether such hybrid states occur in endometrial cancer (EC), however, remains unresolved. Most existing studies rely on in vitro models to explore EMT-related biomarkers and therapeutic targets, yet investigation of EMT dynamics in EC remains limited[5]. A deeper understanding of these transitional states will illuminate the molecular basis of EC metastasis and provide experimental evidence to support the development of novel diagnostic and therapeutic approaches. Cancer-associated fibroblasts (CAFs) are abundant stromal cells in solid tumors, distinct from normal fibroblasts, and play a critical role in cancer progression, metastasis dissemination, and treatment resistance through multifaceted mechanisms [6]. CAFs exhibit remarkable heterogeneity in origin, deriving from resident fibroblasts, mesenchymal stem cells, endothelial cells, or epithelial cells via epithelial–mesenchymal transition (EMT)[6,7]. By secreting of TGF-β, extracellular matrix (ECM) components, and other mediators, CAFs dynamically remodel the tumor microenvironment (TME), facilitating EMT and invasive phenotypes [8]. In EC, CAF-derived signaling activates TGF-β, Wnt, and Notch pathways, directly enhancing tumor cell migration and invasion capacities [9,10]. Nevertheless, the specific CAF subsets driving EMT remain inadequately classified, and their molecular interactions with malignant cells require systematic characterization [7]. Recent single-cell transcriptomics (scRNA-seq) have revealed CAF subsets (e.g., inflammatory CAFs [iCAFs], myofibroblastic CAFs [myCAFs]) across malignancies including EC, each displaying unique transcriptional signatures and differential impacts on tumor progression and immune evasion[7,11]. These findings highlight the need to precisely define CAF heterogeneity in EC, with the goal of selectively targeting pro-tumorigenic subsets while preserving those with antitumor functions. Single-cell RNA sequencing (scRNA-seq) has transformed the resolution of cellular heterogeneity and dynamic molecular processes in complex biological systems, including tumors. scRNA-seq is able to resolve individual cells, capturing rare subpopulations (E/M cells) and depicting dynamic transcriptional states that are critical for processes such as EMT[12–14]. This technology has deconvoluted the tumor microenvironment (TME), mapping diverse malignant, stromal, and immune subsets with discrete functions in tumorigenesis, metastasis, and treatment response [14,15]. Applied to EC, scRNA-seq has elucidated CAFs heterogeneity and immunosuppressive immune cell networks within the TME [15,16] . Critically, this approach provides the necessary framework to dissect the intrinsic heterogeneity and dynamic continuum of EMT states within EC[17,18]. This approach is fundamental to uncovering the complex molecular networks that drive EMT and reveal the core regulatory circuits driving malignant progression. Although prior single-cell RNA sequencing (scRNA-seq) studies in endometrial cancer have advanced our understanding of tumor origins[19], microenvironment heterogeneity[18,20], and potential therapeutic targets[21,22], the dynamic continuum of epithelial-mesenchymal transition (EMT) states and core regulatory molecules driving EMT-driven metastasis remain incompletely characterized[5,11]. Specifically, hybridE/M transitional phenotypes and their spatiotemporal dynamics in clinical specimens, as well as CAF-tumor crosstalk mediators orchestrating malignant progression, require systematic investigation. To address these gaps, we leveraged scRNA-seq of human EC tissues to map EMT trajectories and identify hybrid E/M populations; uncover master regulators of EMT plasticity and establish their functional roles in metastasis and prognostic stratification. Materials and Methods Study design and data collection scRNA-seq datasets from seven EC patients were obtained from the GEO (https://www.ncbi.nlm.nih.gov/geo, Accession: GSE173682 and GSE251923). After quality control, a final gene expression matrix of 52591 high-quality cells was generated by using the R package Seurat (version 5.0.0). Bulk RNA-seq data and matched clinical records for an independent cohort of 521 EC patients were sourced from TCGA-UCEC project. Unsupervised clustering and annotation Batch effects were corrected using Harmony. Dimensionality reduction was performed via principal component analysis (PCA) and Uniform Manifold Approximation and Projection (UMAP). Cell clusters were identified using the Louvain algorithm at a resolution of 0.2 and annotated by marker gene expression. Gene set score The EMT activity of cell clusters was quantified using the AddModuleScore function in R package Seurat with Hallmark EMT gene set (H: Hallmark_EPITHELIAL_MESENCHYMAL_TRANSITION, MSigDB) containing 200 genes (Supplementary Table S2). Gene enrichment analysis Differentially expressed genes (DEGs) were identified via FindAllMarkers in Seurat using Wilcoxon. Human gene sets from 50 hallmark pathways were retrieved from the msigdbr R package (version 7.5.1). Gene Set Variation Analysis (GSVA) was then performed using the GSVA R package (version 1.50.0). Gene Ontology (GO) enrichment analysis and Kyoto Encyclopedia of Genes and Genomes Enrichment (KEGG) Analysis was conducted using the clusterProfiler R package (version 4.10.1). Pseudotime trajectory inference To investigate the pseudotime trajectory analysis of epithelial cells and fibroblasts, we employed the Monocle 2.0 package (version 2.30.0) for scRNA-seq data. The DDRTree algorithm was used for dimensionality reduction while constructing trajectories capable of capturing potential cell developmental or transition states. To visualize gene expression patterns across pseudotime, we applied the plot_genes_in_pseudotime function.The plot_pseudotime_heatmap function was used to generate a pseudotime heatmap, illustrating the dynamic changes in gene expression along the inferred trajectory. cNMF To gain deeper insights into the molecular characteristics of the EMT process, we performed non-negative matrix factorization (NMF) analysis using the GeneNMF package (version 0.4.0). The analysis was conducted with the following parameters: nprograms = 4 and max.genes = 200. Subsequently, we performed enrichment analysis on the obtained modules using the HALLMARK gene sets from the Molecular Signatures Database (MSigDB). Survival analysis We performed survival analysis using TCGA-UCEC cohort data (n=521). The 47-gene EMT signature (Supplementary Table S5) was scored by averaging z-score normalized expression values. Patients were stratified into EMT_high and EMT_low groups. Kaplan-Meier survival curves and Cox regression analyses were conducted via survival package (version 3.5-7) and survminer packages (version 0.4.9). Cell culture The KLE cell line was obtained from the Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences. Cells were maintained in DMEM/F-12 medium (Vivacell) supplemented with 15% fetal bovine serum (Vivacell) at 37°C in a 5% CO₂ humidified incubator. All experiments were performed using cells between passages 3-12 to ensure phenotypic stability. Cell transfection KLE cells were transfected with 50 nM target siRNA (Sangon Biotech) using Lipofectamine 3000 Transfection Kit (Invitrogen) following the manufacturer's protocol. A non-targeting siRNA (siNC) was used as negative control in parallel. All siRNA sequences are listed in Supplementary Table S7. RT-qPCR Total RNA of cell was isolated using RNAeasy Animal RNA isolation Kit (Beyotime) following the manufacturer's instructions, and was then reverse-transcribed into cDNA using a RT-PCR kit (Takara). The RT-PCR assays were performed using SYBR Green Premix Ex Taq on LightCycle. The primers were obtained from Sangon Biotech, and the sequences were listed in Supplementary table S6. The relative expression level of mRNA was calculated according to the 2−ΔΔCt method with GAPDH as an internal control. Western blotting Total protein was isolated from cultured KLE cells using RIPA lysis buffer supplemented with 1% phenylmethylsulfonyl fluoride (PMSF) and 1% protease inhibitor cocktail and the concentration was determined via the A280 absorbance method using a NanoDrop 2000 spectrophotometer. Antibodies for immunoblotting were listed in Supplementary table S7. Wound healing assay Wound healing assays were performed to assess the migratory potential of siRNA transfected KLE cells. A consistent linear wound was generated in confluent cell monolayers, and cell migration was monitored continuously over a 48-hour period under serum-free conditions to eliminate serum-induced chemotaxis interference. The relative wound closure rate was quantified using the formula: [(Area_0h - Area_48h)/Area_0h] × 100%, with data compiled from three independent experimental replicates to ensure reproducibility. Transwell assay The Transwell migration and invasion assesses were respectively used to evaluate the migration and invasion abilities of KLE cells after transfection. For migration tests, Transwell chambers without Matrigel coating were used to evaluate unobstructed cell movement; To conduct the invasion test, a chamber pre-coated with matrix was used to simulate the extracellular matrix (ECM) barrier in vivo. In both experiments, siRNA-transfected KLE cells were inoculated into the upper chamber containing serum-free medium, while the lower chamber was supplemented with complete medium as a chemical attractant to induce the directional movement of the cells. After a 24-hour incubation period, migratory cells and infiltrating cells were fixed, stained, and counted in five randomly selected fields per well. Cell proliferation assay Cell proliferation was measured by the CCK-8 assay. Briefly, KLE cells, after transfecting with siRNA, were seeded in 96-well plates (2×10³cells per well). Every 24 hours, add 10 μL of CCK-8 reagent (MEC) to each well to each well and incubate at 37 ° C for 3 hours. The absorbance was measured at 450 nm. Growth curves were plotted using the measured OD values normalized to day 0. Immunofluorescence Endometrial carcinoma tissue samples were obtained from Daping Hospital, Army Medical University, Chongqing, China, between March 2021 and May 2023. The study protocol involving human tissue samples was approved by the Ethics Committee of Daping Hospital. For retrospective samples, informed consent was exempted due to the anonymization of patient information and the retrospective study design. All procedures were performed in compliance with the Declaration of Helsinki and institutional ethical regulations. Sections underwent deparaffinization/rehydration, antigen retrieval, and blocking. Primary antibodies (Supplementary Table S7) were incubated overnight at 4°C, followed by fluorescence-labeled secondary antibodies (Alexa Fluor 448- and Alexa Fluor 647-conjugated) and DAPI counterstaining. Images were acquired using a digital slide scanner (3D Histech), with fluorescence intensity quantified via ImageJ. Immunohistochemistry After identical deparaffinization/rehydration, sections were incubated with primary antibodies (Supplementary Table S7) overnight at 4°C. Signal detection used DAB chromogen with hematoxylin counterstain. Slides were scanned (Hamamatsu NanoZoomer), and mean optical density (OD) of target proteins was measured in ≥5 tumor fields/region using ImageJ. Results Single-cell morphology and the EMT heterogeneity of EC. To characterize the heterogeneity of EMT in EC, we collected the primary tumor tissues from seven EC patients for scRNA-seq. (Fig.1A, Supplementary Table S1). After initial quality control procedures, we obtained single-cell transcriptomes for a total of 52591 cells across all samples. At a resolution of 0.2, we identified 11 unique cell clusters that were visualized using Unified Manifold Approximation and Projection (UMAP) to facilitate system reduction and cluster visualization (Fig.1B-D). The clusters were annotated according to the expression of classical marker genes as follows: T cells (CD2, CD3D, CD3E, CD3G, TRAC), Myeloid cells ( C1QA, LYZ, CD14, FCGR3A), Endothelial cells (VWF, PECAM1, CLDN5, CDH5, FLT1), Fibroblasts (LUM, COL6A2, DCN, COL1A1), Epithelial cells (EPCAM, CDH1, KRT7, KRT8), B cells (CD79A, CD79B), and Mast cell (KIT) (Fig.1E). Among the seven major cell lines, epithelial cells and fibroblasts accounted for the largest proportions, highlighting their important contributions to EC in the TME. Notably, the cellular composition varied significantly across individual patients (Fig. 1F, Fig. S1A), reflecting the inherent interpatient heterogeneity of EC and underlining the clinical imperative for personalized therapeutic approaches. To further dissect the landscape of EMT in EC, we assessed the EMT scores of different EC cell types using AddModuleScore function (Supplementary Table S2). Consistent with the functional characteristics of fibroblasts in promoting extracellular matrix remodeling and tumor invasion, fibroblasts exhibited the highest EMT scores among all cell types (Fig.1H, Fig. S1B). To validate these findings at the protein level, immunofluorescence staining was performed, which not only confirmed the enhanced presence of EMT in EC tissues compared to normal endometrial tissues but also identified cells in intermediate EMT states (Fig.1G). Collectively, these results indicated that fibroblasts were the primary contributors to EMT signatures in the tumor microenvironment, suggesting a critical role of EMT in EC progression and metastasis. The epithelial cells subsets highly associated with EMT. To characterize the EMT heterogeneity within EC, we performed reclustering of 18,327 epithelial cells isolated from our scRNA-seq dataset. These cells resolved into five distinct subpopulations (epi1-epi5) visualized t as shown by UMAP (Fig.2A,2B). The proportion of each epithelial subpopulation varied significantly, reflecting the substantial heterogeneity within the tumor epithelial compartment (Fig.2C). By comparing the marker genes and signaling pathways of the epithelial cell subgroups, we observed differential expressions of marker genes and signaling pathways, indicating heterogeneity within the epithelial cell subpopulations (Fig.2D). Through GSVA enrichment analysis and EMT scoring on individual cells, we found that epi4 exhibited specific and strong expression of EMT core genes, including COL1A1, TIMP1, SPARC and TPM1, accompanied by significant enrichment of the EMT signaling pathway (Fig.2F,2G). Additionally, subsets epi1 and epi5 demonstrated moderate EMT scores, as well as enrichment of pathways functionally linked to EMT, such as the Wnt/β-catenin signaling-a pathway that has been shown to to promote tumor metastasis and EMT progression. In contrast, epi2 and epi3 maintained the epithelial phenotypes with minimal EMT pathway enrichment (Fig.2F,2G). Based on the EMT score distribution, we stratified the epithelial cells into two functional groups: highEMT_epi (encompassing epi1, epi4, and epi5) and lowEMT_epi (comprising epi2 and epi3) (Fig.2H). Interestingly, we found that highEMT_epi exhibited high levels of angiogenesis and immune invasion (Fig.2H). Together, these findings indicate that epi1, epi4, and epi5 are EMT-high epithelial subsets with pro-metastatic potential. The fibroblasts subsets highly associated with EMT. Our previous analysis demonstrated that among tumor mesenchymal cells, fibroblasts were the primary contributors to EMT (Fig. 1H). To dissect the functional heterogeneity of CAFs, we performed clustering analysis of 17,973 CAFs isolated from 7 EC samples. Based on distinct marker gene expression profiles and UMAP reduction clustering, these CAFs were classified into five subpopulations: MMP3_CAFs, CXCL14_CAFs, PCP4_CAFs, RGS5_CAFs, and EIF5A_CAFs (Fig. 3A, 3B, 3E). To delineate the biological functions of each CAF subpopulation, we conducted GO and GSVA enrichment analysis. CXCL14_CAFs were prominently characterized by extracellular matrix (ECM) related functional signatures, including extracellular matrix organization, extracellular structure organization, and collagen fibril organization (Fig. 3D). GSVA analysis further indicated that EMT and TGFβ signaling pathways were enriched in this subgroup (Fig. 3F). Interestingly, CXCL14_CAFs exhibited high expression of genes such as COL15A1 and PDGFRA. Given the functional and gene expression profiles of CXCL14_CAFs, which align with the characteristics of myofibroblastic CAFs (myCAFs), we propose that CXCL14_CAFs can be classified as myCAFs[23]. PCP4_CAFs were functionally associated with smooth muscle, including smooth muscle contraction and muscle system process (Fig. 3D). GSVA analysis further linked this subpopulation to EMT, Wnt signaling, and myogenesis pathways (Fig. 3F), with Wnt signaling being well-documented to promote EMT and tumor metastatic dissemination. In contrast, MMP3_CAFs were predominantly enriched in nucleic acid metabolism and stress response, such as ribosome biogenesis, rRNA processing and response to oxidative stress, and GSVA analysis also showed that stress response was enriched in this subgroup (Fig. 3D,3F). RGS5_CAFs displayed strong links to stress response and protein folding. EIF5A_CAFs, on the other hand, were associated with with p53 pathway and response to fibroblast growth factor (Fig. S2). To further validate the relationship between CAF subpopulations and EMT, we calculated the EMT scores. Results showed that CXCL14_CAFs and PCP4_CAFs exhibited the highest EMT scores, accompanied by elevated expression of EMT hallmark genes including COL1A1, LUM, TPM2 and TIMP1 (Fig. 3G). Intriguingly, we observed that the proportions of CXCL14_CAFs and PCP4_CAFs were significantly higher in patient 6 and patient 7, both of them were postoperative recurrence patients. (Fig.3C). In conclusion, our findings indicated that CXCL14_CAFs and PCP4_CAFs are fibroblast subsets closely related to EMT, which are conducive to tumor progression and metastasis. The dynamic evolution of EMT in EC To delineate the precise dynamic transition features of EMT in EC, we incorporated these subgroups highly correlated with EMT, including highEMT_epi, CXCL14_CAFs and PCP4_CAFs, and included the lowEMT_epi subgroup to gain deeper insights into this dynamic process (Fig.4A). We performed trajectory inference analysis to visualize the potential cell transition trajectories. Strikingly, our data uncovered a potential progression where lowEMT_epi could evolve into highEMT_epi and subsequently transition into CAFs (Fig.4B,4C). This trajectory delineates a continuous, dynamic EMT evolution process in EC, wherein epithelial cells gradually acquire mesenchymal and CAF-like phenotypes; consistent with this transition pattern, the expression of epithelial marker genes (e.g., EPCAM, CDH1, KRT8, KRT18) gradually decreased along the trajectory, while mesenchymal marker genes (e.g., DCN, ACTA2, COL1A1, VIM) exhibited a steady upward trend (Fig. 4D). This reciprocal expression pattern of epithelial and mesenchymal markers further validates the gradual EMT transition process. To further characterize the molecular features of each stage in the EMT trajectory, we conducted an unbiased trajectory inference analysis using the top 500 genes identified via differentialGeneTest (Supplementary Table S3). The results demonstrated that all genes could be divided into 3 distinct clusters, each corresponding to a different stage of the EMT process. Cluster 3 was identified as an epithelial cluster expressing epithelial marker genes likes KRT7 and KRT8 and enrichment analysis showed that it was enriched in typical functional pathways of epithelial cells (e.g., epithelial cell development, epithelial cell proliferation) (Fig. 4E). Cluster 1 was defined as a mesenchymal cluster with high expression of classical mesenchymal and CAF markers such as COL3A1, FN1, ACTA2, INHBA, POSTN, and SPARC. Enrichment analysis revealed that this cluster was enriched in core mesenchymal pathways (e.g., extracellular matrix organization, focal adhesion, ECM-receptor interaction) and key signaling pathways regulating EMT progression, including PI3K-Akt, MAPK, and Wnt signaling pathways (Fig. 4E). Notably, we identified an intermediate module (cluster2) that corresponded to a "hybrid E/M transition stage" in which cells express a unique set of EMT-related genes (e.g. MMP7, VIM, SPP1, SFRP4, and CD44) (Fig. 4E). Enrichment analysis showed that these cells retained partial epithelial features (e.g., tight junction formation, cell adhesion molecule (CAM) interaction) while acquiring prominent mesenchymal characteristics (e.g., focal adhesion, extracellular structure organization). Importantly, this module was enriched in EMT-regulating signaling pathways such as PI3K-Akt, TNF/NF-κB, and AGE-RAGE, indicating that cells at this stage were actively undergoing EMT reprogramming (Fig. 4E). Taken together, our trajectory and gene clustering analyses revealed a continuous, sequential EMT transition cascade in EC. Moreover, we were the first to characterize the hybrid E/M cells as a critical intermediate stage in this evolution, which bridges epithelial and mesenchymal phenotypes and drives EMT progression through core signaling pathways. This dynamic model provides a novel mechanistic understanding of how EMT contributes to EC progression and metastasis and provides potential therapeutic targets for blocking the EMT cascade. Identification and Clinical Validation of INHBA/POSTN as Core EMT related Genes in EC To dissect the gene expression patterns underlying EMT in EC, we employed the cNMF approach to analyze the EMT high subgroups: PCP4_CAFs, CXCL14_CAFs, and highEMT_epi. This analysis yielded three distinct functional modules with specialized biological roles (Fig.5A, Supplementary Table S4). Module2 was linked to tumor microenvironment adaptation and differentiation, encompassing pathways such as hypoxia and complement activation. While module3 was associated with tumor inflammation regulation with function like inflammatory response, TGF-β signaling and IL6 JAK STAT3 signaling. featuring enrichment of hallmark pathways such as EMT, angiogenesis, and hypoxia—consistent with the core biological processes driving EC aggressiveness (Fig.5A, Supplementary TableS4). By integrating genes from module1, we constructed a comprehensive EC-specific EMT gene signature comprising 47 key genes (Supplementary TableS5). To assess the clinical relevance of this signature, we performed survival analysis using bulk RNA-seq data and clinical records of 521 EC patients from TCGA-UCEC cohort. According to the expression profile of 47 gene signatures, patients were stratified into EMT-low expression group and EMT-high expression group. We found that the EMT-high group exhibited significantly poorer overall survival compared to the EMT-low group (Fig.5B). This finding validates that EMT status, serves as a robust prognostic biomarker for EC patients, highlighting the clinical significance of EMT in disease progression. To pinpoint the core genes driving EMT in EC, we performed an integrated analysis. We first compared scRNA-seq data from 7 EC specimens with 11 normal endometrial samples (Supplementary TableS1, Fig. S3), identifying 8403 upregulated genes in tumor epithelium and 3974 elevated genes in CAFs (avg_log2FC > 0, and p_val_adj < 0.05) (Supplementary TableS6). Through intersection analysis—overlaying these tumor-enriched genes with our 47-gene signature and the top 500 genes from unbiased trajectory inference—two genes, INHBA and POSTN, emerged as the most compelling core candidates (Fig. 5C). Notably, both genes showed stage-dependent expression patterns mirroring the classical EMT markerACTA2 [24], with progressive upregulation along the EMT pseudotime trajectory (Fig. 5D). This dynamic expression profile suggests that INHBA and POSTN may act as key molecular drivers facilitating the sequential EMT transition in EC. To validate these transcriptomic findings at protein level, we performed immunohistochemical (IHC) staining on tissue sections from normal endometrium, early-stage EC (stages Ⅰ-Ⅱ), and late-stage EC (stages Ⅲ-Ⅳ). The protein expression levels of INHBA and POSTN were quantified by ImageJ software, with staining intensity measured as Mean values (Fig.5E,5F). The statistical analysis of IHC results confirmed two key observations: First, compared with normal endometrial tissues, the protein levels of INHBA andPOSTN were significantly upregulated in both early-stage and late-stage EC; Second, their expression was further substantially elevated in late-stage EC relative to early-stage (Fig.5E,5F). The discovery suggest that the upregulation ofINHBA and POSTN are a consistent, stage-dependent event in EC progression. In conclusion, our analyses identified INHBA and POSTN as core EMT-related genes, positioned them not only as reliable biomarkers for EMT progression and clinical outcome, but also as promising therapeutic targets to block EMT-driven EC metastasis and improve patient survival. Functional Validation of INHBA and POSTN in Regulating Endometrial Cancer Cell Malignancy and EMT via In Vitro Assays To dissect the functional roles of INHBA and POSTN in driving EC progression, we performed targeted gene silencing experiments using specific small interfering RNAs (siRNAs) in the KLE EC cell line. qRT-PCR and Western blot analyses were first conducted to validate the efficiency of siRNA-mediated knockdown (KD). As shown in Fig. 6A and 6B, transfection with INHBA-targeting or POSTN-targeting siRNAs resulted in a significant reduction in both mRNA and protein levels compared to the non-targeting control siRNA (siNC) group. Among the three siRNA candidates tested for each gene, siINHBA#3 and siPOSTN#3 exhibited the most robust KD efficiency, and thus were selected for subsequent functional assays. Then transwell migration and invasion assays were performed. The results demonstrated that knockdown of either INHBA or POSTN led to a marked decrease in both migration and invasion capabilities of KLE cells (Fig. 6C). These findings were further corroborated by wound-healing assay. At 48 h post-wounding, the wound closure rate in siINHBA and siPOSTN groups was significantly lower than that in the siNC group (Fig. 6D,6E), confirming that loss of INHBA or POSTN impairs EC cell migratory activity. To evaluate the impact ofINHBA or POSTNKD on EC cell malignant properties, we assessed cell proliferation via CCK-8 assay. The results showed that compared with the siNC group, both siINHBA and siPOSTN groups exhibited a significant decrease in cell proliferation capacity at all time points, with the most prominent reduction observed at 96h (Fig. 6F). This indicated that INHBA and POSTN also contribute to EC cell proliferation, a key hallmark of cancer cell malignancy. Given the established link between EMT and cancer cell metastasis, we next investigated whether INHBA and POSTN regulate the EMT process in KLE cells by detecting the expression of core EMT-associated proteins via Western blot. As shown in Fig. 6G, knockdown of INHBA or POSTN induced a clear reversal of the EMT phenotype in KLE cells: the expression of E-cadherin was significantly upregulated, while the expression of N-cadherin was markedly downregulated compared to the siNC group. Collectively, these in vitro functional assays demonstrate that INHBA and POSTN play critical pro-oncogenic roles in EC by promoting cell migration, invasion, and EMT progression—further supporting their potential as therapeutic targets for inhibiting EC metastasis. Discussion Our study delineates EMT heterogeneity and dynamics in EC through scRNA-seq, uncovering novel cellular trajectories and core regulatory mechanisms. We identified fibroblasts as pivotal contributors to EMT，consistent with their established role in tumor progression[6,9]. Beyond conventional fibroblast populations, we resolved distinct EMT-high epithelial subsets and functionally divergent CAF subpopulations. Pseudotime trajectory analysis revealed a sequential EMT progression axis: lowEMT_epi → highEMT_epi → CAFs, directly capturing the continuum of transitional states—a previously uncharacterized feature in EC. Within this axis, we identified a hybrid E/M state co-expressing epithelial (CD24 and MUC1) and mesenchymal (VIM and SPP1) markers, with functional enrichment in PI3K-Akt and TNF/NF-κB pathways. This hybrid E/M state acts as a metastatic hub, retainsing partial epithelial traits (enabling cell survival) while acquiring mesenchymal properties (facilitating dissemination). This finding addresses a major gap in current EC research. By mapping this trajectory and intermediate state directly in clinical specimens, our work provides the first in vivo evidence of a continuous EMT cascade in EC, bridging epithelial and stromal phenotypes to drive tumor progression. To identify EMT related genes, we preformed cNMF module analysis and identified 47 genes. The 47-gene EC-specific EMT signature stratified TCGA-UCEC patients into EMT-high and EMT-low groups with significantly different overall survival (p < 0.0001), reinforcing EMT as a robust prognostic biomarker for EC. This signature’s clinical utility lies in its potential to guide adjuvant therapy decisions, as high EMT scores may identify patients at elevated risk of recurrence who would benefit from intensified treatment. Additionally,we identifiedINHBA and POSTN as the central regulators of EMT in EC. Both genes exhibited stage-dependent upregulation along with the EMT trajectory and progression from normal endometrium to advanced EC, validating their role as stage-specific biomarkers. INHBA, a member of the TGF-β superfamily, encodes the inhibin-βA subunit that forms functional dimers for transducing downstream signals[25]. It is well known that the TGF-β pathway has an environmentally dependent duality in cancer, acting as a tumor suppressor in the early stages and as a cancer-promoting driver in advanced disease [25,26]. Our findings extend recent reports linking INHBA to malignant progression in ovarian [27] and esophageal cancers [28], demonstrating its role in promoting EMT, migration, and invasion in EC. Notably, INHBA generates an immunosuppressive TME by modulating cancer-associated fibroblast function [27] and induces resistance to PD-L1 blockade by suppressing IFN-γ signaling[29]. This suggests that INHBA may regulate EMT and immune evasion in EC, which have dual effects and implications for combination therapy. POSTN is an ECM protein that has emerged as a key regulator of EMT and TME remodeling across malignancies [30,31]. Mechanistically, POSTN activates the ILK-Akt pathway to drive EMT in EC[32].Our functional analysis confirms that its knockdown can reverse EMT (upregulates E-cadherin, downregulates N-cadherin) and impairscell proliferation. Consistent with studies in hepatocellular carcinoma [30] and ovarian cancer [31], POSTN+ CAFs in EC may remodel the ECM to create a pro-tumorigenic niche that supports immune suppression and metastatic dissemination. Clinically, POSTN is associated with poor outcomes in multiple cancers [33,34], and our IHC data validate its progressive upregulation from normal endometrium to late-stage EC, positioning it as a stage-specific biomarker for EC progression. EMT-driven cellular plasticity and lineage transitions are conserved across epithelial malignancies, but exhibit cancer-specific patterns. In colorectal cancer, Yang[14] identified BHLHE40 as a transcriptional driver of EMT, with malignant epithelial cells transitioning into CXCL1+ CAFs. In esophageal squamous cell carcinoma, Guo[16] reported EMT initiation in precancerous lesions, peaking before lymph node metastasis. In contrast, our study reveals a steady increase in EMT activity from early-stage to late-stage EC, with no evidence of precancerous EMT initiation or pre-metastatic peaks. Such context-dependent disparities emphasize the need for cancer-specific EMT investigations—while cross-tumor comparisons reveal conserved mechanisms, direct extrapolation risks overlooking key disease-specific features [3,4]. Emerging evidence highlights EMT as a promising therapeutic target in EC. Lengrand et al. [35]demonstrated that pharmacological blockade of netrin-1 with NP137 reduces EMT-positive tumor cells and enhances chemotherapy sensitivity, and Cassier et al. [36]validated these findings in EC mouse models and a first-in-human clinical trial—confirming the translational potential of EMT inhibition. Our identification of INHBA and POSTN as core EMT drivers provides novel therapeutic targets: inhibiting these genes could intercept the EMT cascade, suppress metastasis, and potentially sensitize EC to existing therapies. For example, targeting INHBA may disrupt both EMT and immunosuppression[27,29], while POSTN inhibition could block ECM remodeling and CAF-tumor crosstalk[30,31]—strategies worthy of preclinical and clinical exploration. There are several limitations to this study. First, all single-cell data are derived from stage I EC samples. This limits the comprehensive assessment of the advanced EMT characteristics of metastatic tumor niches. Second, we did not validate EMT dynamics in a mouse model of EC. Third, although we confirmed the role of INHBA and POSTN in driving EMT, we did not conduct a detailed study of upstream regulatory factors and downstream influencing factors. Finally, our functional analysis relied exclusively on gene knockdown approaches, resulting in the potential effects of overexpression being unexplored. Conclusion Our study provides a comprehensive characterization of EMT heterogeneity and dynamic progression in EC, identifies CXCL14_CAFs/PCP4_CAFs as key EMT-associated stromal subsets, uncovers a hybrid E/M intermediate state, and validates INHBA and POSTN as core EMT regulators with prognostic and therapeutic potential. These findings deepen our understanding of EC biology, fill critical gaps in current EMT research [5,11]. As EMT inhibition emerges as a viable therapeutic approach in EC[36,37], our identification of INHBA and POSTN as central drivers offers promising targets to improve outcomes for patients with advanced or recurrent disease. Declarations Acknowledgments We would also like to acknowledge the TCGA and GEO databases for offering publicly available datasets that were used in our analysis.We would like to thank all the teachers in the Daping Hospital, Army Medical University for their support. Author contributions All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by DL, BW and JH. The first draft of the manuscript was written by DL, YM and XC all authors commented on previous versions of the manuscript. Funding The authors declare that financial support was received for the research of this article. This study was supported by the Basic Research Project of Army Medical University (Grant No.50122-4039). Data availability The scRNA-seq datasets generated in this study are available in the GEO ( https://www.ncbi.nlm.nih.gov/geo , Accession: GSE173682 and GSE251923). Bulk RNA-seq data and clinical records were retrieved from The Cancer Genome Atlas (TCGA-UCEC project, https://portal.gdc.cancer.gov). All datasets used in our study are from previously published studies, and detailed information can be found in the Supplementary Material. Further inquiries can be directed to the corresponding authors. Conflict of interest The authors declare no competing interests. Ethical approval The studies involving humans were approved by Ethics Committee of Daping Hospital，Army Medical University (Approval No.202131). The studies were conducted in accordance with the local legislation and institutional requirements. References Crosbie EJ, Kitson SJ, McAlpine JN, MukhopadhyayA, Powell ME, Singh N. Endometrial cancer. Lancet. 2022;399:1412–28.https://doi.org/10.1016/S0140-6736(22)00323-3 Corr BR, Erickson BK, Barber EL, Fisher CM, Slomovitz B. Advances in the management of endometrial cancer. BMJ. 2025;388:e080978. https://doi.org/10.1136/bmj-2024-080978 Yang J, Antin P, Berx G, Blanpain C, Brabletz T, Bronner M, et al. Guidelines and definitions for research on epithelial-mesenchymal transition. Nat Rev Mol Cell Biol. 2020;21:341–52. https://doi.org/10.1038/s41580-020-0237-9 Akhmetkaliyev A, Alibrahim N, Shafiee D,Tulchinsky E. EMT/MET plasticity in cancer and Go-or-Grow decisions inquiescence: the two sides of the same coin? Mol Cancer. 2023;22:90.https://doi.org/10.1186/s12943-023-01793-z Ren F, Wang L, Wang Y, Wang J, Wang Y, Song X, et al. Single-cell transcriptome profiles the heterogeneity of tumor cells and microenvironments for different pathological endometrial cancer and identifies specific sensitive drugs. Cell Death Dis. 2024;15:571. https://doi.org/10.1038/s41419-024-06960-8 Kalluri R. The biology and function of fibroblasts in cancer. Nat Rev Cancer. 2016;16:582–98. https://doi.org/10.1038/nrc.2016.73 Yamamoto Y, Kasashima H, Fukui Y, Tsujio G, Yashiro M, Maeda K. The heterogeneity of cancer-associated fibroblast subpopulations: Their origins, biomarkers, and roles in the tumor microenvironment. Cancer Sci. 2023;114:16–24. https://doi.org/10.1111/cas.15609 Cheng PSW, Zaccaria M, Biffi G. Functional heterogeneity of fibroblasts in primary tumors and metastases. Trends Cancer. 2024;S2405-8033(24)00260-7. https://doi.org/10.1016/j.trecan.2024.11.005 Biffi G, Tuveson DA. Diversity and Biology of Cancer-Associated Fibroblasts. Physiol Rev. 2021;101:147–76. https://doi.org/10.1152/physrev.00048.2019 Cords L, de Souza N, Bodenmiller B. Classifying cancer-associated fibroblasts-The good, the bad, and the target. Cancer Cell. 2024;42:1480–5. https://doi.org/10.1016/j.ccell.2024.08.011 Yu Z, Zhang J, Zhang Q, Wei S, Shi R, Zhao R, et al. Single-cell sequencing reveals the heterogeneity and intratumoral crosstalk in human endometrial cancer. Cell Proliferat. 2022;55:e13249. https://doi.org/10.1111/cpr.13249 Stuart T, Satija R. Integrative single-cell analysis. Nat Rev Genet. 2019;20:257–72. https://doi.org/10.1038/s41576-019-0093-7 Papalexi E, Satija R. Single-cell RNA sequencing to explore immune cell heterogeneity. Nat Rev Immunol. 2018;18:35–45. https://doi.org/10.1038/nri.2017.76 Yang S, Zhang D, Sun Q, Nie H, Zhang Y, Wang X,et al. Single-Cell and Spatial Transcriptome Profiling Identifies theTranscription Factor BHLHE40 as a Driver of EMT in Metastatic ColorectalCancer. Cancer Res. 2024;84:2202–17. https://doi.org/10.1158/0008-5472.CAN-23-3264 Wu Q, Jiang G, Sun Y, Li B. Reanalysis of single-cell data reveals macrophage subsets associated with the immunotherapy response and prognosis of patients with endometrial cancer. Exp Cell Res. 2023;430:113736. https://doi.org/10.1016/j.yexcr.2023.113736 Guo D, Sheng K, Zhang Q, Li P, Sun H, Wang Y, et al. Single-cell transcriptomic analysis reveals the landscape of epithelial-mesenchymal transition molecular heterogeneity in esophageal squamous cell carcinoma. Cancer Lett. 2024;587:216723. https://doi.org/10.1016/j.canlet.2024.216723 Wang S, Sun S-T, Zhang X-Y, Ding H-R, Yuan Y, He J-J, et al. The Evolution of Single-Cell RNA Sequencing Technology and Application: Progress and Perspectives. Int J Mol Sci. 2023;24:2943. https://doi.org/10.3390/ijms24032943 González-Martínez S, Pérez-Mies B, Cortés J, Palacios J. Single-cell RNA sequencing in endometrial cancer: exploring the epithelial cells and the microenvironment landscape. Front Immunol. 2024;15:1425212. https://doi.org/10.3389/fimmu.2024.1425212 Ren X, Liang J, Zhang Y, Jiang N, Xu Y, Qiu M, et al. Single-cell transcriptomic analysis highlights origin and pathological process of human endometrioid endometrial carcinoma. Nat Commun. 2022;13:6300. https://doi.org/10.1038/s41467-022-33982-7 Hashimoto S, Tabuchi Y, Yurino H, Hirohashi Y, Deshimaru S, Asano T, et al. Comprehensive single-cell transcriptome analysis reveals heterogeneity in endometrioid adenocarcinoma tissues. Sci Rep-uk. 2017;7:14225. https://doi.org/10.1038/s41598-017-14676-3 Wang J, Meng F, Mao F. Single cell sequencing analysis and transcriptome analysis constructed the liquid-liquid phase separation(LLPS)-related prognostic model for endometrial cancer. Front Oncol. 2022;12:1005472. https://doi.org/10.3389/fonc.2022.1005472 Cochrane DR, Campbell KR, Greening K, Ho GC, Hopkins J, Bui M, et al. Single cell transcriptomes of normal endometrial derived organoids uncover novel cell type markers and cryptic differentiation of primary tumours. J Pathol. 2020;252:201–14. https://doi.org/10.1002/path.5511 Chhabra Y, Weeraratna AT. Fibroblasts in cancer: Unity in heterogeneity. Cell. 2023;186:1580–609. https://doi.org/10.1016/j.cell.2023.03.016 Brabletz S, Schuhwerk H, Brabletz T, Stemmler MP. Dynamic EMT: a multi-tool for tumor progression. EMBO J. 2021;40:e108647. https://doi.org/10.15252/embj.2021108647 Deng Z, Fan T, Xiao C, Tian H, Zheng Y, Li C, et al. TGF-β signaling in health, disease, and therapeutics. Signal Transduct Target Ther. 2024;9:61. https://doi.org/10.1038/s41392-024-01764-w Colak S, Ten Dijke P. Targeting TGF-β Signaling in Cancer. Trends Cancer. 2017;3:56–71. https://doi.org/10.1016/j.trecan.2016.11.008 Hu Y, Recouvreux MS, Haro M, Taylan E, Taylor-Harding B, Walts AE, et al. INHBA(+) cancer-associated fibroblasts generate an immunosuppressive tumor microenvironment in ovarian cancer. NPJ Precis Oncol. 2024;8:35. https://doi.org/10.1038/s41698-024-00523-y Wang J-J, Chen D-X, Zhang Y, Xu X, Cai Y, Wei W-Q, et al. Elevated expression of the RNA-binding protein IGF2BP1 enhances the mRNA stability of INHBA to promote the invasion and migration of esophageal squamous cancer cells. Exp Hematol Oncol. 2023;12:75. https://doi.org/10.1186/s40164-023-00429-8 Li F, Gu L, Tong Y, Chen R, Chen S, Yu X, et al. INHBA promotes tumor growth and induces resistance to PD-L1 blockade by suppressing IFN-γ signaling. Acta Pharmacol Sin. Nature Publishing Group; 2025;46:448–61. https://doi.org/10.1038/s41401-024-01381-x Wang H, Liang Y, Liu Z, Zhang R, Chao J, Wang M, et al. POSTN+ cancer-associated fibroblasts determine the efficacy of immunotherapy in hepatocellular carcinoma. J Immunother Cancer. 2024;12:e008721. https://doi.org/10.1136/jitc-2023-008721 Lin S-C, Liao Y-C, Chen P-M, Yang Y-Y, Wang Y-H, Tung S-L, et al. Periostin promotes ovarian cancer metastasis by enhancing M2 macrophages and cancer-associated fibroblasts via integrin-mediated NF-κB and TGF-β2 signaling. J Biomed Sci. 2022;29:109. https://doi.org/10.1186/s12929-022-00888-x Zheng Q-M, Lu J-J, Zhao J, Wei X, Wang L, Liu P-S. Periostin Facilitates the Epithelial-Mesenchymal Transition of Endometrial Epithelial Cells through ILK-Akt Signaling Pathway. Biomed Res Int. 2016;2016:9842619. https://doi.org/10.1155/2016/9842619 De Oliveira Macena Y, Cezar MEN, Lira CBF, De Oliveira LBDM, Almeida TN, Costa ADAV, et al. The roles of periostin derived from cancer-associated fibroblasts in tumor progression and treatment response. Cancer Metastasis Rev. 2024;44:11. https://doi.org/10.1007/s10555-024-10233-3 Null JL, Kim DJ, McCann JV, Pramoonjago P, Fox JW, Zeng J, et al. Periostin+ Stromal Cells Guide Lymphovascular Invasion by Cancer Cells. Cancer Res. 2023;83:2105–22. https://doi.org/10.1158/0008-5472.CAN-22-2412 Lengrand J, Pastushenko I, Vanuytven S, Song Y, Venet D, Sarate RM, et al. Pharmacological targeting of netrin-1 inhibits EMT in cancer. Nature. 2023;620:402–8. https://doi.org/10.1038/s41586-023-06372-2 Cassier PA, Navaridas R, Bellina M, Rama N, Ducarouge B, Hernandez-Vargas H, et al. Netrin-1 blockade inhibits tumour growth and EMT features in endometrial cancer. Nature. 2023;620:409–16. https://doi.org/10.1038/s41586-023-06367-z Xia X, Yin K, Wang S. Targeting of netrin-1 by monoclonal antibody NP137 inhibits the EMT in cancer. J Immunother Cancer. 2024;12:e008937. https://doi.org/10.1136/jitc-2024-008937 Additional Declarations No competing interests reported. Supplementary Files tableS2.csv tableS7.xlsx tableS3.csv tableS1.xlsx tableS4.csv tableS5.csv SupplementaryMaterial.pdf tableS6.xlsx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8267986","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":582026621,"identity":"7f222a38-80cc-4d36-acb6-70e371c0b208","order_by":0,"name":"Dan Liu","email":"","orcid":"","institution":"Daping Hospital, Army Medical University","correspondingAuthor":false,"prefix":"","firstName":"Dan","middleName":"","lastName":"Liu","suffix":""},{"id":582026622,"identity":"231b8e3e-d2e1-4854-bcc5-d82ddbd5cb30","order_by":1,"name":"Ben Wang","email":"","orcid":"","institution":"Daping Hospital, Army Medical University","correspondingAuthor":false,"prefix":"","firstName":"Ben","middleName":"","lastName":"Wang","suffix":""},{"id":582026623,"identity":"2dbd0dac-32ce-476f-8919-23d8405456ec","order_by":2,"name":"Jie Huang","email":"","orcid":"","institution":"Daping Hospital, Army Medical University","correspondingAuthor":false,"prefix":"","firstName":"Jie","middleName":"","lastName":"Huang","suffix":""},{"id":582026624,"identity":"dad8fc12-6b1e-437b-b868-215beb4df7f4","order_by":3,"name":"Ge Diao","email":"","orcid":"","institution":"Daping Hospital, Army Medical University","correspondingAuthor":false,"prefix":"","firstName":"Ge","middleName":"","lastName":"Diao","suffix":""},{"id":582026625,"identity":"7711543e-af88-4de6-8556-06587932c1c9","order_by":4,"name":"Yunhe Ma","email":"","orcid":"","institution":"Daping Hospital, Army Medical University","correspondingAuthor":false,"prefix":"","firstName":"Yunhe","middleName":"","lastName":"Ma","suffix":""},{"id":582026626,"identity":"d02f1ee0-f4ef-4b8a-8ccc-9127124a1a26","order_by":5,"name":"Xuena Chen","email":"","orcid":"","institution":"Chongqing University","correspondingAuthor":false,"prefix":"","firstName":"Xuena","middleName":"","lastName":"Chen","suffix":""},{"id":582026627,"identity":"f334fb60-2245-4093-a503-149a6ee2a348","order_by":6,"name":"Runbo Li","email":"","orcid":"","institution":"Daping Hospital, Army Medical University","correspondingAuthor":false,"prefix":"","firstName":"Runbo","middleName":"","lastName":"Li","suffix":""},{"id":582026628,"identity":"5e64a04b-2f71-400a-84b8-2d7c5c9b9ec0","order_by":7,"name":"XianHua Wang","email":"","orcid":"","institution":"Daping Hospital, Army Medical University","correspondingAuthor":false,"prefix":"","firstName":"XianHua","middleName":"","lastName":"Wang","suffix":""},{"id":582026629,"identity":"6f150adf-7b89-4d7b-b8cd-999b9935a170","order_by":8,"name":"Jianxin Guo","email":"","orcid":"","institution":"Daping Hospital, Army Medical University","correspondingAuthor":false,"prefix":"","firstName":"Jianxin","middleName":"","lastName":"Guo","suffix":""},{"id":582026630,"identity":"2fd96648-6d0a-4c67-88ae-a742b6cec72c","order_by":9,"name":"Jian Han","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA0klEQVRIiWNgGAWjYBACA4YDDMwgBj9UgLGBaC2SbcRrYYBoMThGrBZzxgOMjwt+Hc4zvt+dupmHwUZ2wwHmZw/wabFsOMBsPLPvcLHZMd5tt3kY0ow3HGAzN8DrsAMH2KR5ew4nboNoOZy44QAPmwRRWja3gbX8J1ILzw+g4WxgLQeI0gL0S0N64oxjudtuzjFINp55mM0Mv5YboBD7Y53Y33x22403FXayfcebn+HVwiBx/gMDY1szzAQGaDThA/wNQOJPHSFlo2AUjIJRMJIBAMGcUSrC3O1AAAAAAElFTkSuQmCC","orcid":"","institution":"Daping Hospital, Army Medical University","correspondingAuthor":true,"prefix":"","firstName":"Jian","middleName":"","lastName":"Han","suffix":""}],"badges":[],"createdAt":"2025-12-03 08:39:06","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8267986/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8267986/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":101785420,"identity":"f33dcf64-5721-4e6f-a0c5-dd3bb1dc3503","added_by":"auto","created_at":"2026-02-03 15:36:29","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":435440,"visible":true,"origin":"","legend":"\u003cp\u003eSingle-cell landscape and the EMT heterogeneity in EC\u003cstrong\u003e A\u003c/strong\u003e Schematic overview of the study design. \u003cstrong\u003eB-D\u003c/strong\u003e UMAP plot of single-cell transcriptomes: \u003cstrong\u003eB\u003c/strong\u003e coded by annotated cell types. \u003cstrong\u003eC\u003c/strong\u003e ratified by patient origin. \u003cstrong\u003eD\u003c/strong\u003e Segregated by unsupervised clustering. \u003cstrong\u003eE\u003c/strong\u003e The dotplot of the representative marker genes for the cell types. \u003cstrong\u003eF\u003c/strong\u003e Stacked bar plot showing the proportional distribution of 7 major cell types across individual patients. \u003cstrong\u003eG\u003c/strong\u003e Representative immunofluorescence double-staining images of E-cadherin (green) and N-cadherin (red) in normal l endometrial tissue and endometrial cancer tissue. Scale bar,20μm. \u003cstrong\u003eH\u003c/strong\u003e UMAP plot of EMT score level across 7 cell types.\u003c/p\u003e","description":"","filename":"figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-8267986/v1/7f9ffc4564e200bc4e17e64a.png"},{"id":101785428,"identity":"cd24188a-248b-4074-b763-6e70e90f87ac","added_by":"auto","created_at":"2026-02-03 15:36:29","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":5713,"visible":true,"origin":"","legend":"\u003cp\u003eThe epithelial cells subsets highly associated with EMT\u003cstrong\u003e A-B\u003c/strong\u003e UMAP plot of the epithelial cells: \u003cstrong\u003eA\u003c/strong\u003e colored by epithelium subtypes. \u003cstrong\u003eB\u003c/strong\u003eStratified by patient origin. \u003cstrong\u003eC\u003c/strong\u003e Pie charts show the proportions of the epithelium subtypes. \u003cstrong\u003eD\u003c/strong\u003e The dotplot displaying the top 5 marker genes from the differential analysis of five epithelial subtypes.\u003cstrong\u003e E\u003c/strong\u003e The violin plot showing the EMT score level of epithelium subtypes. \u003cstrong\u003eF\u003c/strong\u003e Heatmap of GSVA enrichment score for hallmarker pathways across epithelium subtype. \u003cstrong\u003eG\u003c/strong\u003e The dot plot of some classic EMT genes for each subtypes. \u003cstrong\u003eH\u003c/strong\u003e The violin plot comparing EMT score, angiogenesis score, and immune invasion score between the highEMT_epi and lowEMT_epi groups.\u003c/p\u003e","description":"","filename":"placeholderimage.png","url":"https://assets-eu.researchsquare.com/files/rs-8267986/v1/94e54d40614e1c96a88d6e42.png"},{"id":101785423,"identity":"1f3b9514-5374-479f-89ae-ccdc14e95be2","added_by":"auto","created_at":"2026-02-03 15:36:29","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":382872,"visible":true,"origin":"","legend":"\u003cp\u003eThe fibroblasts subsets highly associated with EMT\u003cstrong\u003e A-B\u003c/strong\u003e UMAP visualization of fibroblast subpopulations: \u003cstrong\u003eA\u003c/strong\u003e Color-coded by five distinct CAF subtypes. \u003cstrong\u003eB\u003c/strong\u003e Stratified by patient origin. \u003cstrong\u003eC\u003c/strong\u003e Stacked bar plot showing the proportional distribution of the fibroblast subtypes across individual patients. \u003cstrong\u003eD\u003c/strong\u003e GO enrichment analysis of biological processes. \u003cstrong\u003eE\u003c/strong\u003e Heatmap of top 10 marker genes used for fibroblast subpopulation annotation.\u003cstrong\u003e F\u003c/strong\u003e Heatmap of GSVA enrichment score for hallmarker pathways across fibroblast subtypes. \u003cstrong\u003eG\u003c/strong\u003e Violin plot showing the EMT score level of fibroblast subtypes, the dot plot of some classic EMT genes for each subtype.\u003c/p\u003e","description":"","filename":"figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-8267986/v1/81febc2302c8d57b3172251f.png"},{"id":101785421,"identity":"a92e8af5-f6e2-4394-80d9-89e9ba1ab82a","added_by":"auto","created_at":"2026-02-03 15:36:29","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":1151262,"visible":true,"origin":"","legend":"\u003cp\u003eThe dynamic evolution of EMT in EC \u003cstrong\u003eA\u003c/strong\u003eUMAP visualization of EMT subpopulation, including lowEMT_epi, highEMT_epi, CXCL14_CAFs and PCP4_CAFs. \u003cstrong\u003eB-C\u003c/strong\u003ePseudotime trajectory analysis of EMT progression: \u003cstrong\u003eB\u003c/strong\u003e Cell distribution along trajectory colored by subpopulation type. \u003cstrong\u003eC\u003c/strong\u003e Trajectory branches colored by inferred pseudotime. \u003cstrong\u003eD\u003c/strong\u003e Temporal variation in epithelial marker (\u003cem\u003eEPCAM, CDH1andKRT8\u003c/em\u003e) and mesenchymal marker (\u003cem\u003eDCN, ACTA2 and COL1A1\u003c/em\u003e). \u003cstrong\u003eE\u003c/strong\u003ePseudotemporal gene expression patterns and corresponding pathway activation: Heatmap showing gene expression clusters during EMT transition (Left). Enrichment anlysis associated with each transitional phase (Right).\u003c/p\u003e","description":"","filename":"figure4.png","url":"https://assets-eu.researchsquare.com/files/rs-8267986/v1/706e39607322294148f8a248.png"},{"id":101880794,"identity":"bb65793a-a382-426a-9ae5-a55d9030af0e","added_by":"auto","created_at":"2026-02-04 15:06:18","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":686386,"visible":true,"origin":"","legend":"\u003cp\u003eIdentification and Clinical Validation of INHBA/POSTN as Core EMT-Associated Genes in EC\u003cstrong\u003e A\u003c/strong\u003e The heatmap illustrates the pairwise correlation patterns among 3 modules, with the hallmark pathways associated with each module labeled in the right-hand panel. \u003cstrong\u003eB\u003c/strong\u003eThe KM plots showing the OS of the EMT_high and the EMT_low groups in TCGA dataset. \u003cstrong\u003eC\u003c/strong\u003e Venn diagram illustrating the intersection of four gene sets: genes highly expressed in tumor epithelial cells, genes highly expressed in tumor fibroblasts, top 500 genes from pseudotime trajectory analysis, and 47 genes screened via cNMF. \u003cstrong\u003eD\u003c/strong\u003e Temporal variation in \u003cem\u003eINHBA\u003c/em\u003e and \u003cem\u003ePOSTN\u003c/em\u003e.\u003cstrong\u003e E\u003c/strong\u003e Representative IHC images of \u003cem\u003eINHBA\u003c/em\u003eand \u003cem\u003ePOSTN\u003c/em\u003e in normal and tumor samples with early and late stages. Scale bars, 100μm. \u003cstrong\u003eF\u003c/strong\u003e Histogram plots showed the IHC staining intensity of \u003cem\u003eINHBA\u003c/em\u003eand \u003cem\u003ePOSTN\u003c/em\u003e. **, P \u0026lt; 0.01; ***, P \u0026lt; 0.001.\u003c/p\u003e","description":"","filename":"figure5.png","url":"https://assets-eu.researchsquare.com/files/rs-8267986/v1/82f0e75894af63ad8fd7cada.png"},{"id":101785426,"identity":"0c1f9d97-34d8-484f-b17d-2e55a5770047","added_by":"auto","created_at":"2026-02-03 15:36:29","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":601248,"visible":true,"origin":"","legend":"\u003cp\u003eFunctional Validation of INHBA and POSTN in Regulating Endometrial Cancer Cell Malignancy and EMT via In Vitro Assays\u003cstrong\u003e A\u003c/strong\u003e qRT-PCR analysis of KLE cells after transfecting specific siRNA compared with siNC. \u003cstrong\u003eB\u003c/strong\u003e Western blot analysis of KLE cells after transfecting specific siRNA compared with siNC (above). The column diagram shows the protein relative expression level of INHBA and POSTN\u003cem\u003e \u003c/em\u003e(below). \u003cstrong\u003eC\u003c/strong\u003e Transwell migration and invasion assay in INHBA KD and POSTN KD cells (left). Scale bar,100μm. Histogram plots showed the cell number (right). \u003cstrong\u003eD\u003c/strong\u003eThe wound healing assay in INHBA KD and POSTN KD cells. \u003cstrong\u003eE\u003c/strong\u003e Histogram plots showed the wound closure. \u003cstrong\u003eF\u003c/strong\u003e CCK-8 assay was used to detect the proliferative capacity in INHBA KD and POSTN KD cells. \u003cstrong\u003eG\u003c/strong\u003e WB illustrated the expression of crucial proteins of EMT in INHBA KD and POSTN KD cells(left). The column diagram shows the protein relative expression level(right). All data were presented as the means ± SD. *, P \u0026lt; 0.05; **, P \u0026lt; 0.01; ***, P \u0026lt; 0.001.\u003c/p\u003e","description":"","filename":"figure6.png","url":"https://assets-eu.researchsquare.com/files/rs-8267986/v1/77090809fe08ff8f6cc44522.png"},{"id":109057077,"identity":"e0d8e74f-6dc0-47e6-911d-e8fba9943e6f","added_by":"auto","created_at":"2026-05-12 07:46:34","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3120714,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8267986/v1/39e9a65f-0627-47ed-8048-5fc9db5ec5b8.pdf"},{"id":101785419,"identity":"08c259dc-db0b-4d39-ad03-85087d91c1df","added_by":"auto","created_at":"2026-02-03 15:36:29","extension":"csv","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":10739,"visible":true,"origin":"","legend":"","description":"","filename":"tableS2.csv","url":"https://assets-eu.researchsquare.com/files/rs-8267986/v1/d3cf12323d981ee13a31bb90.csv"},{"id":101785424,"identity":"38db984e-adba-4351-bf72-668de9b00532","added_by":"auto","created_at":"2026-02-03 15:36:29","extension":"xlsx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":12695,"visible":true,"origin":"","legend":"","description":"","filename":"tableS7.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-8267986/v1/42cbe211eb3d42df55a80db3.xlsx"},{"id":101785430,"identity":"82245b14-9169-4cdc-8b2e-3aaf4c8781f7","added_by":"auto","created_at":"2026-02-03 15:36:29","extension":"csv","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":7240,"visible":true,"origin":"","legend":"","description":"","filename":"tableS3.csv","url":"https://assets-eu.researchsquare.com/files/rs-8267986/v1/df35672330bb9634d5477c78.csv"},{"id":101785427,"identity":"7684395d-45ca-4d7d-a832-849d67c63139","added_by":"auto","created_at":"2026-02-03 15:36:29","extension":"xlsx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":11828,"visible":true,"origin":"","legend":"","description":"","filename":"tableS1.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-8267986/v1/d0e76004e970bf7f25dd3a31.xlsx"},{"id":101785436,"identity":"96d8e36e-e6a4-4cb5-964d-d15e1cd720f7","added_by":"auto","created_at":"2026-02-03 15:36:32","extension":"csv","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":4177,"visible":true,"origin":"","legend":"","description":"","filename":"tableS4.csv","url":"https://assets-eu.researchsquare.com/files/rs-8267986/v1/0e1c2c8e7aa923ccc6cf1f33.csv"},{"id":101785429,"identity":"bc4f289f-0591-48d6-9bd7-42633f4e5e7c","added_by":"auto","created_at":"2026-02-03 15:36:29","extension":"csv","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":487,"visible":true,"origin":"","legend":"","description":"","filename":"tableS5.csv","url":"https://assets-eu.researchsquare.com/files/rs-8267986/v1/360ecb7910f05e2699c2c832.csv"},{"id":101785431,"identity":"cc4b1404-1c1a-4b3c-a7bc-32d9b1e4ed32","added_by":"auto","created_at":"2026-02-03 15:36:29","extension":"pdf","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":535475,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryMaterial.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8267986/v1/293ce8298474b371587aa844.pdf"},{"id":101785432,"identity":"a4830ccb-80d5-4e40-9125-1d8366e1c475","added_by":"auto","created_at":"2026-02-03 15:36:29","extension":"xlsx","order_by":7,"title":"","display":"","copyAsset":false,"role":"supplement","size":996755,"visible":true,"origin":"","legend":"","description":"","filename":"tableS6.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-8267986/v1/fdd0db1323d1851e8bbae1a4.xlsx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Single-Cell analysis Uncovers the Dynamic Process of EMT in Endometrial Cancer and the Core Roles of INHBA and POSTN in Regulating Tumor Malignant Phenotypes","fulltext":[{"header":"Introduction","content":"\u003cp\u003eEndometrial cancer (EC) has emerged as a leading malignancy of the female reproductive tract and represents an escalating global health burden. As highlighted in recent reviews, the rising incidence of EC is closely linked to two major drivers: the global aging population and the obesity epidemic, which have driven a steady increase in new diagnoses over the past decade [1,2]. Clinically, approximately two thirds of patients are diagnosed at an early-stage, and exhibit a favorable prognoses [1]. However, the prognosis diverges sharply for patients with metastatic or recurrent EC: despite advances in systemic treatments, including chemotherapy and targeted therapy, survival rates remain below 20%, highlighting the limitations of current treatment strategies [2]. This critical gap between early-stage success and advanced disease failure has prompted an urgent demand to identify molecular biomarkers that can predict risk of metastasis, stratify prognosis, and guide targeted interventions—efforts that are central to addressing the unmet needs of EC patients.\u003c/p\u003e\n\u003cp\u003eEpithelial–mesenchymal transition (EMT) is a biological process in which epithelial cells acquire mesenchymal phenotypes through a series of coordinated steps. During EMT, epithelial cells progressively lose defining features such as E-cadherin and keratin expression, while gaining mesenchymal markers including N-cadherin and vimentin. EMT plays diverse roles in embryonic development, wound healing, tissue regeneration, and tumor progression. In cancer, it is increasingly recognised that EMT underpins metastasis, stemness, and drug resistance[3]. Rather than a binary shift, EMT is now understood as a dynamic continuum, during which hybrid E/M cells emerge that exhibit both epithelial and mesenchymal traits, thereby facilitating dissemination and colonisation[4]. Whether such hybrid states occur in endometrial cancer (EC), however, remains unresolved. Most existing studies rely on in vitro models to explore EMT-related biomarkers and therapeutic targets, yet investigation of EMT dynamics in EC remains limited[5]. A deeper understanding of these transitional states will illuminate the molecular basis of EC metastasis and provide experimental evidence to support the development of novel diagnostic and therapeutic approaches.\u003c/p\u003e\n\u003cp\u003eCancer-associated fibroblasts (CAFs) are abundant stromal cells in solid tumors, distinct from normal fibroblasts, and play a critical role in cancer progression, metastasis dissemination, and treatment resistance through multifaceted mechanisms [6]. CAFs exhibit remarkable heterogeneity in origin, deriving from resident fibroblasts, mesenchymal stem cells, endothelial cells, or epithelial cells via epithelial–mesenchymal transition (EMT)[6,7]. By secreting of TGF-β, extracellular matrix (ECM) components, and other mediators, CAFs dynamically remodel the tumor microenvironment (TME), facilitating EMT and invasive phenotypes [8]. In EC, CAF-derived signaling activates TGF-β, Wnt, and Notch pathways, directly enhancing tumor cell migration and invasion capacities [9,10]. Nevertheless, the specific CAF subsets driving EMT remain inadequately classified, and their molecular interactions with malignant cells require systematic characterization [7]. Recent single-cell transcriptomics (scRNA-seq) have revealed CAF subsets (e.g., inflammatory CAFs [iCAFs], myofibroblastic CAFs [myCAFs]) across malignancies including EC, each displaying unique transcriptional signatures and differential impacts on tumor progression and immune evasion[7,11]. These findings highlight the need to precisely define CAF heterogeneity in EC, with the goal of selectively targeting pro-tumorigenic subsets while preserving those with antitumor functions.\u003c/p\u003e\n\u003cp\u003eSingle-cell RNA sequencing (scRNA-seq) has transformed the resolution of cellular heterogeneity and dynamic molecular processes in complex biological systems, including tumors. scRNA-seq is able to resolve individual cells, capturing rare subpopulations (E/M cells) and depicting dynamic transcriptional states that are critical for processes such as EMT[12–14]. This technology has deconvoluted the tumor microenvironment (TME), mapping diverse malignant, stromal, and immune subsets with discrete functions in tumorigenesis, metastasis, and treatment response [14,15]. Applied to EC, scRNA-seq has elucidated CAFs heterogeneity and immunosuppressive immune cell networks within the TME [15,16] . Critically, this approach provides the necessary framework to dissect the intrinsic heterogeneity and dynamic continuum of EMT states within EC[17,18]. This approach is fundamental to uncovering the complex molecular networks that drive EMT and reveal the core regulatory circuits driving malignant progression.\u003c/p\u003e\n\u003cp\u003eAlthough prior single-cell RNA sequencing (scRNA-seq) studies in endometrial cancer have advanced our understanding of tumor origins[19], microenvironment heterogeneity[18,20], and potential therapeutic targets[21,22], the dynamic continuum of epithelial-mesenchymal transition (EMT) states and core regulatory molecules driving EMT-driven metastasis remain incompletely characterized[5,11]. Specifically, hybridE/M transitional phenotypes and their spatiotemporal dynamics in clinical specimens, as well as CAF-tumor crosstalk mediators orchestrating malignant progression, require systematic investigation. To address these gaps, we leveraged scRNA-seq of human EC tissues to map EMT trajectories and identify hybrid E/M populations; uncover master regulators of EMT plasticity and establish their functional roles in metastasis and prognostic stratification.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003ch2\u003eStudy design and data collection\u0026nbsp;\u003c/h2\u003e\n\u003cp\u003escRNA-seq datasets from seven EC patients were obtained from the GEO (https://www.ncbi.nlm.nih.gov/geo, Accession: GSE173682 and GSE251923). After quality control, a final gene expression matrix of 52591 high-quality cells was generated by using the R package Seurat (version 5.0.0). Bulk RNA-seq data and matched clinical records for an independent cohort of 521 EC patients were sourced from TCGA-UCEC project.\u003c/p\u003e\n\u003ch2\u003eUnsupervised clustering and annotation\u003c/h2\u003e\n\u003cp\u003eBatch effects were corrected using Harmony. Dimensionality reduction was performed via principal component analysis (PCA) and Uniform Manifold Approximation and Projection (UMAP). Cell clusters were identified using the Louvain algorithm at a resolution of 0.2 and\u0026nbsp;annotated by marker gene expression.\u003c/p\u003e\n\u003ch1\u003eGene set score\u003c/h1\u003e\n\u003cp\u003eThe EMT activity of cell clusters was quantified using the AddModuleScore function in R package Seurat with Hallmark EMT gene set (H: Hallmark_EPITHELIAL_MESENCHYMAL_TRANSITION, MSigDB) containing 200 genes (Supplementary Table S2).\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003e\u0026nbsp;Gene enrichment analysis\u003c/h2\u003e\n\u003cp\u003eDifferentially expressed genes (DEGs) were identified via FindAllMarkers in Seurat using Wilcoxon. Human gene sets from 50 hallmark pathways were retrieved from the msigdbr R package (version 7.5.1). Gene Set Variation Analysis (GSVA) was then performed using the GSVA R package (version 1.50.0). Gene Ontology (GO) enrichment analysis and Kyoto Encyclopedia of Genes and Genomes Enrichment (KEGG) Analysis was conducted using the clusterProfiler R package (version 4.10.1).\u003c/p\u003e\n\u003ch2\u003e\u0026nbsp;Pseudotime trajectory inference\u003c/h2\u003e\n\u003cp\u003eTo investigate the pseudotime trajectory analysis of epithelial cells and fibroblasts, we employed the Monocle 2.0 package (version 2.30.0) for scRNA-seq data. The DDRTree algorithm was used for dimensionality reduction while constructing trajectories capable of capturing potential cell developmental or transition states. To visualize gene expression patterns across pseudotime, we applied the plot_genes_in_pseudotime function.The plot_pseudotime_heatmap function was used to generate a pseudotime heatmap, illustrating the dynamic changes in gene expression along the inferred trajectory.\u003c/p\u003e\n\u003ch2\u003ecNMF\u003c/h2\u003e\n\u003cp\u003eTo gain deeper insights into the molecular characteristics of the EMT process, we performed non-negative matrix factorization (NMF) analysis using the GeneNMF package (version 0.4.0). The analysis was conducted with the following parameters: nprograms = 4 and max.genes = 200. Subsequently, we performed enrichment analysis on the obtained modules using the HALLMARK gene sets from the Molecular Signatures Database (MSigDB).\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003e\u0026nbsp;Survival analysis\u003c/h2\u003e\n\u003cp\u003eWe performed survival analysis using TCGA-UCEC cohort data (n=521). The 47-gene EMT signature (Supplementary Table S5)\u0026nbsp;was scored by averaging z-score normalized expression values. Patients were stratified into EMT_high and EMT_low groups. Kaplan-Meier survival curves and Cox regression analyses were conducted via survival package (version 3.5-7) and survminer packages\u0026nbsp;(version\u0026nbsp;0.4.9).\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eCell culture\u003c/h2\u003e\n\u003cp\u003eThe KLE cell line was obtained from the Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences. Cells were maintained in DMEM/F-12 medium (Vivacell) supplemented with 15% fetal bovine serum (Vivacell) at 37\u0026deg;C in a 5% CO₂ humidified incubator. All experiments were performed using cells between passages 3-12 to ensure phenotypic stability.\u003c/p\u003e\n\u003ch2\u003eCell transfection\u003c/h2\u003e\n\u003cp\u003eKLE cells were transfected with 50 nM target siRNA (Sangon Biotech) using Lipofectamine 3000 Transfection Kit (Invitrogen) following the manufacturer\u0026apos;s protocol. A non-targeting siRNA (siNC) was used as negative control in parallel. All siRNA sequences are listed in Supplementary Table S7.\u003c/p\u003e\n\u003ch2\u003eRT-qPCR\u003c/h2\u003e\n\u003cp\u003eTotal RNA of cell was isolated using RNAeasy Animal RNA isolation Kit (Beyotime) following the manufacturer\u0026apos;s instructions, and was then reverse-transcribed into cDNA using a RT-PCR kit (Takara). The RT-PCR assays were performed using SYBR Green Premix Ex Taq on LightCycle. The primers were obtained from Sangon Biotech, and the sequences were listed in Supplementary table S6. The relative expression level of mRNA was calculated according to the 2\u0026minus;\u0026Delta;\u0026Delta;Ct method with GAPDH as an internal control.\u003c/p\u003e\n\u003ch2\u003eWestern blotting\u003c/h2\u003e\n\u003cp\u003eTotal protein was isolated from cultured KLE cells using RIPA lysis buffer supplemented with 1% phenylmethylsulfonyl fluoride (PMSF) and 1% protease inhibitor cocktail and the concentration was determined via the A280 absorbance method using a NanoDrop 2000 spectrophotometer. Antibodies for immunoblotting were listed in Supplementary table S7.\u003c/p\u003e\n\u003ch2\u003eWound healing assay\u003c/h2\u003e\n\u003cp\u003eWound healing assays were performed to assess the migratory potential of siRNA transfected KLE cells. A consistent linear wound was generated in confluent cell monolayers, and cell migration was monitored continuously over a 48-hour period under serum-free conditions to eliminate serum-induced chemotaxis interference. The relative wound closure rate was quantified using the formula: [(Area_0h - Area_48h)/Area_0h] \u0026times; 100%, with data compiled from three independent experimental replicates to ensure reproducibility.\u003c/p\u003e\n\u003ch2\u003eTranswell assay\u003c/h2\u003e\n\u003cp\u003eThe Transwell migration and invasion assesses were respectively used to evaluate the migration and invasion abilities of KLE cells after transfection. For migration tests, Transwell chambers without Matrigel coating were used to evaluate unobstructed cell movement; To conduct the invasion test, a chamber pre-coated with matrix was used to simulate the extracellular matrix (ECM) barrier in vivo. In both experiments, siRNA-transfected KLE cells were inoculated into the upper chamber containing serum-free medium, while the lower chamber was supplemented with complete medium as a chemical attractant to induce the directional movement of the cells. After a 24-hour incubation period, migratory cells and infiltrating cells were fixed, stained, and counted in five randomly selected fields per well.\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eCell proliferation assay\u003c/h2\u003e\n\u003cp\u003eCell proliferation was measured by the CCK-8 assay. Briefly, KLE cells, after transfecting with siRNA, were seeded in 96-well plates (2\u0026times;10\u0026sup3;cells per well). Every 24 hours, add 10 \u0026mu;L of CCK-8 reagent (MEC) to each well to each well and incubate at 37 \u0026deg; C for 3 hours. The absorbance was measured at 450 nm. Growth curves were plotted using the measured OD values normalized to day 0.\u003c/p\u003e\n\u003ch2\u003eImmunofluorescence\u0026nbsp;\u003c/h2\u003e\n\u003cp\u003eEndometrial carcinoma tissue samples were obtained from Daping Hospital, Army Medical University, Chongqing, China, between March 2021 and May 2023. The study protocol involving human tissue samples was approved by the Ethics Committee of Daping Hospital. For retrospective samples, informed consent was exempted due to the anonymization of patient information and the retrospective study design. All procedures were performed in compliance with the Declaration of Helsinki and institutional ethical regulations.\u0026nbsp;Sections underwent deparaffinization/rehydration, antigen retrieval, and blocking. Primary antibodies (Supplementary Table S7) were incubated overnight at 4\u0026deg;C, followed by fluorescence-labeled secondary antibodies (Alexa Fluor 448- and Alexa Fluor 647-conjugated) and DAPI counterstaining. Images were acquired using a digital slide scanner (3D Histech), with fluorescence intensity quantified via ImageJ.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eImmunohistochemistry\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;After identical deparaffinization/rehydration, sections were incubated with primary antibodies (Supplementary Table S7) overnight at 4\u0026deg;C. Signal detection used DAB chromogen with hematoxylin counterstain. Slides were scanned (Hamamatsu NanoZoomer), and mean optical density (OD) of target proteins was measured in \u0026ge;5 tumor fields/region using ImageJ.\u003c/p\u003e"},{"header":"Results","content":"\u003ch2\u003eSingle-cell morphology and the EMT heterogeneity\u0026nbsp;of\u0026nbsp;EC.\u003c/h2\u003e\n\u003cp\u003eTo characterize the heterogeneity of EMT in EC, we collected the primary tumor tissues from seven EC patients for scRNA-seq. (Fig.1A, Supplementary Table S1). After initial quality control procedures, we obtained single-cell transcriptomes for a total of 52591 cells across all samples. At a resolution of 0.2, we identified 11 unique cell clusters that were visualized using Unified Manifold Approximation and Projection (UMAP) to facilitate system reduction and cluster visualization (Fig.1B-D). The clusters were annotated according to the expression of classical marker genes as follows: T cells (CD2, CD3D, CD3E, CD3G, TRAC), Myeloid cells \u003cem\u003e(\u003c/em\u003eC1QA, LYZ, CD14, FCGR3A), Endothelial cells (VWF, PECAM1, CLDN5, CDH5, FLT1), Fibroblasts (LUM, COL6A2, DCN, COL1A1), Epithelial cells (EPCAM, CDH1, KRT7, KRT8), B cells (CD79A, CD79B), and Mast cell (KIT) (Fig.1E). Among the seven major cell lines, epithelial cells and fibroblasts accounted for the largest proportions, highlighting their important contributions to EC in the TME. Notably, the cellular composition varied significantly across individual patients (Fig. 1F, Fig. S1A), reflecting the inherent interpatient heterogeneity of EC and underlining the clinical imperative for personalized therapeutic approaches.\u003c/p\u003e\n\u003cp\u003eTo further dissect the landscape of EMT in EC, we assessed the EMT scores of different EC cell types using AddModuleScore function (Supplementary Table S2). Consistent with the functional characteristics of fibroblasts in promoting extracellular matrix remodeling and tumor invasion, fibroblasts exhibited the highest EMT scores among all cell types (Fig.1H, Fig. S1B). To validate these findings at the protein level, immunofluorescence staining was performed, which not only confirmed the enhanced presence of EMT in EC tissues compared to normal endometrial tissues but also identified cells in intermediate EMT states (Fig.1G). Collectively, these results indicated that fibroblasts were the primary contributors to EMT signatures in the tumor microenvironment, suggesting a critical role of EMT in EC progression and metastasis.\u003c/p\u003e\n\u003ch2\u003eThe epithelial cells subsets highly associated with EMT.\u003c/h2\u003e\n\u003cp\u003eTo characterize the EMT heterogeneity within EC, we performed reclustering of 18,327 epithelial cells isolated from our scRNA-seq dataset. These cells resolved into five distinct subpopulations (epi1-epi5) visualized t as shown by UMAP (Fig.2A,2B). The proportion of each epithelial subpopulation varied significantly, reflecting the substantial heterogeneity within the tumor epithelial compartment (Fig.2C).\u003c/p\u003e\n\u003cp\u003eBy comparing the marker genes and signaling pathways of the epithelial cell subgroups, we observed differential expressions of marker genes and signaling pathways, indicating heterogeneity within the epithelial cell subpopulations (Fig.2D). Through GSVA enrichment analysis and EMT scoring on individual cells, we found that epi4 exhibited specific and strong expression of EMT core genes, including COL1A1, TIMP1, SPARC and TPM1, accompanied by significant enrichment of the EMT signaling pathway\u0026nbsp;(Fig.2F,2G).\u0026nbsp;Additionally,\u0026nbsp;subsets epi1 and epi5 demonstrated moderate EMT scores, as well as enrichment of pathways functionally linked to EMT, such as the Wnt/β-catenin signaling-a pathway that has been shown to to promote tumor metastasis and EMT progression. In contrast, epi2 and epi3 maintained the epithelial phenotypes with minimal EMT pathway enrichment\u0026nbsp;(Fig.2F,2G).\u0026nbsp;Based on the EMT score distribution, we stratified the epithelial cells into two functional groups: highEMT_epi (encompassing epi1, epi4, and epi5) and lowEMT_epi (comprising epi2 and epi3)\u0026nbsp;(Fig.2H). Interestingly, we found that highEMT_epi\u0026nbsp;exhibited\u0026nbsp;high levels\u0026nbsp;of\u0026nbsp;angiogenesis and immune invasion (Fig.2H).\u0026nbsp;Together, these findings indicate that epi1, epi4, and epi5 are EMT-high epithelial subsets with pro-metastatic potential.\u003c/p\u003e\n\u003ch2\u003eThe fibroblasts subsets highly associated with EMT.\u003c/h2\u003e\n\u003cp\u003eOur previous analysis demonstrated that among tumor mesenchymal cells, fibroblasts were the primary contributors to EMT (Fig. 1H). To dissect the functional heterogeneity of CAFs, we performed clustering analysis of 17,973 CAFs isolated from 7 EC samples. Based on distinct marker gene expression profiles and UMAP reduction clustering, these CAFs were classified into five subpopulations: MMP3_CAFs, CXCL14_CAFs, PCP4_CAFs, RGS5_CAFs, and EIF5A_CAFs (Fig. 3A, 3B, 3E).\u003c/p\u003e\n\u003cp\u003eTo delineate the biological functions of each CAF subpopulation, we conducted GO and GSVA enrichment analysis. CXCL14_CAFs were prominently characterized by extracellular matrix (ECM) related functional signatures, including extracellular matrix organization, extracellular structure organization, and collagen fibril organization (Fig. 3D). GSVA analysis further indicated that EMT and TGFβ signaling pathways were enriched in this subgroup (Fig. 3F). Interestingly, CXCL14_CAFs exhibited high expression of genes such as COL15A1 and PDGFRA. Given the functional and gene expression profiles of CXCL14_CAFs, which align with the characteristics of myofibroblastic CAFs (myCAFs), we propose that CXCL14_CAFs can be classified as myCAFs[23]. PCP4_CAFs were functionally associated with smooth muscle, including smooth muscle contraction and muscle system process (Fig. 3D). GSVA analysis further linked this subpopulation to EMT, Wnt signaling, and myogenesis pathways (Fig. 3F), with Wnt signaling being well-documented to promote EMT and tumor metastatic dissemination. In contrast, MMP3_CAFs were predominantly enriched in nucleic acid metabolism and stress response, such as ribosome biogenesis, rRNA processing and\u0026nbsp;response to oxidative stress, and GSVA analysis also showed that stress response was enriched in this subgroup (Fig. 3D,3F). RGS5_CAFs displayed strong links to stress response and protein folding. EIF5A_CAFs, on the other hand, were associated with with p53 pathway and response to fibroblast growth factor (Fig. S2).\u003c/p\u003e\n\u003cp\u003eTo further validate the relationship between CAF subpopulations and EMT, we calculated the EMT scores. Results showed that CXCL14_CAFs and PCP4_CAFs exhibited the highest EMT scores, accompanied by elevated expression of EMT hallmark genes including COL1A1, LUM, TPM2 and TIMP1 (Fig. 3G). Intriguingly, we observed that the proportions of CXCL14_CAFs and PCP4_CAFs were significantly higher in patient 6 and patient 7,\u0026nbsp;both of them\u0026nbsp;were postoperative recurrence patients.\u0026nbsp;(Fig.3C).\u0026nbsp;In conclusion, our findings indicated\u0026nbsp;that CXCL14_CAFs and PCP4_CAFs are fibroblast subsets closely related to EMT, which are conducive to tumor progression and metastasis.\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eThe dynamic evolution of EMT in EC\u003c/h2\u003e\n\u003cp\u003eTo delineate the precise dynamic transition features of EMT in EC, we incorporated these subgroups highly correlated with EMT, including highEMT_epi, CXCL14_CAFs and PCP4_CAFs, and included the lowEMT_epi subgroup to gain deeper insights into this dynamic process (Fig.4A).\u0026nbsp;We performed trajectory inference analysis to visualize the potential cell transition trajectories. Strikingly, our data\u0026nbsp;uncovered\u0026nbsp;a potential progression where lowEMT_epi could evolve into highEMT_epi and subsequently transition into CAFs (Fig.4B,4C).\u0026nbsp;This trajectory delineates a continuous, dynamic EMT evolution process in EC, wherein epithelial cells gradually acquire mesenchymal and CAF-like phenotypes; consistent with this transition pattern, the expression of epithelial marker genes (e.g., EPCAM, CDH1, KRT8, KRT18)\u0026nbsp;gradually decreased\u0026nbsp;along the trajectory, while mesenchymal marker genes (e.g., DCN, ACTA2, COL1A1, VIM)\u0026nbsp;exhibited\u0026nbsp;a steady upward trend (Fig. 4D). This reciprocal expression pattern of epithelial and mesenchymal markers further validates the gradual EMT transition process.\u003c/p\u003e\n\u003cp\u003eTo further characterize the molecular features of each stage in the EMT trajectory, we conducted an unbiased trajectory inference analysis using the top 500 genes identified via differentialGeneTest (Supplementary Table S3). The results demonstrated that all genes could be divided into 3 distinct clusters, each corresponding to a different stage of the EMT process. Cluster 3 was identified as an epithelial cluster expressing epithelial marker genes likes KRT7 and KRT8 and enrichment analysis showed that it was enriched in typical functional pathways of epithelial cells (e.g., epithelial cell development, epithelial cell proliferation) (Fig. 4E).\u0026nbsp;Cluster\u0026nbsp;1 was defined as\u0026nbsp;a\u0026nbsp;mesenchymal cluster\u0026nbsp;with high expression of classical mesenchymal and CAF markers such as COL3A1, FN1, ACTA2, INHBA, POSTN, and SPARC.\u0026nbsp;Enrichment analysis revealed that this cluster was enriched in core mesenchymal pathways (e.g., extracellular matrix organization, focal adhesion, ECM-receptor interaction) and key signaling pathways regulating EMT progression, including PI3K-Akt, MAPK, and Wnt signaling pathways (Fig. 4E). Notably, we identified\u0026nbsp;an intermediate module (cluster2) that corresponded to a \"hybrid E/M transition stage\" in which cells express a unique set of EMT-related genes (e.g. MMP7, VIM, SPP1, SFRP4, and CD44) (Fig. 4E). Enrichment analysis showed that these cells retained partial epithelial features (e.g., tight junction formation, cell adhesion molecule (CAM) interaction) while acquiring prominent mesenchymal characteristics (e.g., focal adhesion, extracellular structure organization). Importantly, this module was enriched in EMT-regulating signaling pathways such as PI3K-Akt, TNF/NF-κB, and AGE-RAGE, indicating that cells at this stage were actively undergoing EMT reprogramming (Fig. 4E).\u003c/p\u003e\n\u003cp\u003eTaken together, our trajectory and gene clustering analyses revealed a continuous, sequential EMT transition cascade in EC. Moreover, we were the first to characterize the hybrid E/M cells as a critical intermediate stage in this evolution, which bridges epithelial and mesenchymal phenotypes and drives EMT progression through core signaling pathways. This dynamic model provides a novel mechanistic understanding of how EMT contributes to EC progression and metastasis and provides potential therapeutic targets for blocking the EMT cascade.\u003c/p\u003e\n\u003ch2\u003eIdentification and Clinical Validation of INHBA/POSTN as Core EMT\u0026nbsp;related\u0026nbsp;Genes in EC\u003c/h2\u003e\n\u003cp\u003eTo dissect the gene expression patterns underlying EMT in EC, we employed the cNMF approach to analyze the EMT high subgroups: PCP4_CAFs, CXCL14_CAFs, and highEMT_epi. This analysis yielded three distinct functional modules with specialized biological roles (Fig.5A, Supplementary Table S4). Module2 was linked to tumor microenvironment adaptation and differentiation, encompassing pathways such as hypoxia and complement activation. While module3 was associated with tumor\u0026nbsp;inflammation\u0026nbsp;regulation\u0026nbsp;with function like\u0026nbsp;inflammatory response,\u0026nbsp;TGF-β\u0026nbsp;signaling and\u0026nbsp;IL6 JAK STAT3 signaling.\u0026nbsp;featuring enrichment of hallmark pathways such as EMT, angiogenesis, and hypoxia—consistent with the core biological processes driving EC aggressiveness\u0026nbsp;(Fig.5A, Supplementary TableS4).\u0026nbsp;By integrating genes from module1, we constructed a comprehensive EC-specific EMT gene signature comprising 47 key genes\u0026nbsp;(Supplementary TableS5).\u003c/p\u003e\n\u003cp\u003eTo assess the clinical relevance of this signature, we performed survival analysis using bulk RNA-seq data and clinical records of 521 EC patients from TCGA-UCEC cohort. According to the expression profile of 47 gene signatures, patients were stratified into EMT-low expression group and EMT-high expression group. We found that the EMT-high group exhibited significantly poorer overall survival compared to the EMT-low group\u0026nbsp;(Fig.5B). This finding validates that EMT status, serves as a robust prognostic biomarker for EC patients, highlighting the clinical significance of EMT in disease progression.\u003c/p\u003e\n\u003cp\u003eTo pinpoint the core genes driving EMT in EC, we performed an integrated analysis. We first compared scRNA-seq data from 7 EC specimens with 11 normal endometrial samples (Supplementary TableS1, Fig. S3), identifying 8403 upregulated genes in tumor epithelium and 3974 elevated genes in CAFs (avg_log2FC \u0026gt; 0, and p_val_adj \u0026lt; 0.05) (Supplementary TableS6). Through intersection analysis—overlaying these tumor-enriched genes with our 47-gene signature and the top 500 genes from unbiased trajectory inference—two genes, INHBA and POSTN, emerged as the most compelling core candidates (Fig. 5C). Notably, both genes showed stage-dependent expression patterns mirroring the classical EMT markerACTA2 [24], with progressive upregulation along the EMT pseudotime trajectory (Fig. 5D). This dynamic expression profile suggests\u0026nbsp;that INHBA and POSTN may act as key molecular drivers facilitating the sequential EMT transition in EC.\u003c/p\u003e\n\u003cp\u003eTo validate these transcriptomic findings at protein level, we performed immunohistochemical (IHC) staining on tissue sections from normal endometrium, early-stage EC (stages Ⅰ-Ⅱ), and late-stage EC (stages Ⅲ-Ⅳ). The protein expression levels of INHBA and POSTN were quantified by ImageJ software, with\u0026nbsp;staining intensity measured as Mean values (Fig.5E,5F). The statistical analysis of IHC results confirmed two key observations: First, compared with normal endometrial tissues, the protein levels of INHBA andPOSTN were significantly upregulated in both early-stage and late-stage EC; Second, their expression was further substantially elevated in late-stage EC relative to early-stage (Fig.5E,5F). The discovery suggest\u0026nbsp;that the upregulation ofINHBA and POSTN are a consistent, stage-dependent event in EC progression. In conclusion, our analyses identified INHBA and POSTN as core EMT-related genes, positioned them not only as reliable biomarkers for EMT progression and clinical outcome, but also as promising therapeutic targets to block EMT-driven EC metastasis and improve patient survival.\u003c/p\u003e\n\u003ch1\u003eFunctional Validation of INHBA and POSTN in Regulating Endometrial Cancer Cell Malignancy and EMT via In Vitro Assays\u003c/h1\u003e\n\u003cp\u003eTo dissect the functional roles of INHBA and POSTN in driving EC progression, we performed targeted gene silencing experiments using specific small interfering RNAs (siRNAs) in the KLE EC cell line. qRT-PCR and Western blot analyses were first conducted to validate the efficiency of siRNA-mediated knockdown (KD). As shown in Fig. 6A and 6B, transfection with INHBA-targeting or POSTN-targeting siRNAs resulted in a significant reduction in both mRNA and protein levels compared to the non-targeting control siRNA (siNC) group. Among the three siRNA candidates tested for each gene,\u0026nbsp;siINHBA#3 and siPOSTN#3 exhibited the most robust KD efficiency, and thus were selected for subsequent functional assays.\u003c/p\u003e\n\u003cp\u003eThen transwell migration and invasion assays were performed. The results demonstrated that\u0026nbsp;knockdown of either INHBA or POSTN led to a marked decrease in both migration and invasion capabilities of KLE cells\u0026nbsp;(Fig. 6C). These findings were further corroborated by wound-healing assay. At 48 h post-wounding, the wound closure rate in siINHBA and siPOSTN groups was significantly lower than that in the siNC group (Fig. 6D,6E), confirming that loss of INHBA or POSTN impairs EC cell migratory activity. To evaluate the impact ofINHBA or POSTNKD on EC cell malignant properties, we assessed cell proliferation via CCK-8 assay. The results showed that compared with the siNC group, both siINHBA and siPOSTN groups exhibited a significant decrease in cell proliferation capacity at all time points, with the most prominent reduction observed at 96h (Fig. 6F). This indicated that INHBA and POSTN also contribute to EC cell proliferation, a key hallmark of cancer cell malignancy.\u003c/p\u003e\n\u003cp\u003eGiven the established link between EMT and cancer cell metastasis, we next investigated whether INHBA and POSTN regulate the EMT process in KLE cells by detecting the expression of core EMT-associated proteins via Western blot. As shown in Fig. 6G, knockdown of INHBA or POSTN induced a clear reversal of the EMT phenotype in KLE cells: the expression of E-cadherin was significantly upregulated, while the expression of N-cadherin was markedly downregulated compared to the siNC group. Collectively, these in vitro functional assays demonstrate that INHBA and POSTN play critical pro-oncogenic roles in EC by promoting cell migration, invasion, and EMT progression—further supporting their potential as therapeutic targets for inhibiting EC metastasis.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eOur study delineates EMT heterogeneity and dynamics in EC through scRNA-seq, uncovering novel cellular trajectories and core regulatory mechanisms. We identified fibroblasts as pivotal contributors to EMT，consistent with their established role in tumor progression[6,9]. Beyond conventional fibroblast populations, we resolved distinct EMT-high epithelial subsets and functionally divergent CAF subpopulations.\u0026nbsp;Pseudotime trajectory analysis revealed a sequential EMT progression axis: lowEMT_epi → highEMT_epi → CAFs, directly capturing the continuum of transitional states—a previously uncharacterized feature in EC. Within this axis, we identified a hybrid E/M state co-expressing epithelial (CD24 and MUC1) and mesenchymal (VIM and SPP1)\u0026nbsp;markers, with functional enrichment in PI3K-Akt and TNF/NF-κB pathways. This hybrid E/M state acts as a metastatic hub, retainsing partial epithelial traits (enabling cell survival) while acquiring mesenchymal properties (facilitating dissemination). This finding addresses a major gap in current EC research. By mapping this trajectory and intermediate state directly in clinical specimens, our work provides the first in vivo evidence of a continuous EMT cascade in EC, bridging epithelial and stromal phenotypes to drive tumor progression.\u003c/p\u003e\n\u003cp\u003eTo identify EMT related genes, we preformed cNMF module analysis and identified 47 genes. The 47-gene EC-specific EMT signature stratified TCGA-UCEC patients into EMT-high and EMT-low groups with significantly different overall survival (p \u0026lt; 0.0001), reinforcing EMT as a robust prognostic biomarker for EC. This signature’s clinical utility lies in its potential to guide adjuvant therapy decisions, as high EMT scores may identify patients at elevated risk of recurrence who would benefit from intensified treatment.\u0026nbsp;Additionally,we\u0026nbsp;identifiedINHBA and POSTN as\u0026nbsp;the\u0026nbsp;central regulators of EMT in EC. Both genes exhibited stage-dependent upregulation along\u0026nbsp;with\u0026nbsp;the EMT trajectory and progression from normal endometrium to advanced EC, validating their role as stage-specific biomarkers.\u0026nbsp;INHBA, a member of the TGF-β superfamily,\u0026nbsp;encodes\u0026nbsp;the inhibin-βA subunit\u0026nbsp;that\u0026nbsp;forms functional dimers\u0026nbsp;for\u0026nbsp;transducing\u0026nbsp;downstream signals[25]. It is well known that the TGF-β pathway has an environmentally dependent duality in cancer, acting as a tumor suppressor in the early stages and as a cancer-promoting driver in advanced disease\u0026nbsp;[25,26]. Our findings extend recent reports linking INHBA to malignant progression in ovarian\u0026nbsp;[27]\u0026nbsp;and esophageal cancers\u0026nbsp;[28], demonstrating its role in promoting EMT, migration, and invasion in EC. Notably, INHBA generates\u0026nbsp;an immunosuppressive TME by modulating cancer-associated fibroblast function\u0026nbsp;[27]\u0026nbsp;and induces\u0026nbsp;resistance to PD-L1 blockade by suppressing IFN-γ signaling[29]. This\u0026nbsp;suggests\u0026nbsp;that INHBA may\u0026nbsp;regulate\u0026nbsp;EMT and immune evasion in EC,\u0026nbsp;which\u0026nbsp;have dual effects and implications for combination therapy.\u0026nbsp;POSTN\u0026nbsp;is\u0026nbsp;an ECM protein\u0026nbsp;that\u0026nbsp;has emerged as a key regulator of EMT and TME remodeling across malignancies\u0026nbsp;[30,31]. Mechanistically, POSTN activates the ILK-Akt pathway to drive EMT in EC[32].Our functional analysis confirms\u0026nbsp;that its knockdown can reverse EMT (upregulates E-cadherin, downregulates N-cadherin) and impairscell proliferation. Consistent with studies in hepatocellular carcinoma\u0026nbsp;[30]\u0026nbsp;and ovarian cancer\u0026nbsp;[31], POSTN+ CAFs in EC may remodel the ECM to create a pro-tumorigenic niche\u0026nbsp;that\u0026nbsp;supports\u0026nbsp;immune suppression and metastatic dissemination. Clinically, \u003cem\u003ePOSTN\u003c/em\u003e is associated with poor outcomes in multiple cancers\u0026nbsp;[33,34], and our IHC data validate its progressive upregulation from normal endometrium to late-stage EC, positioning it as a stage-specific biomarker for EC progression.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eEMT-driven cellular plasticity and lineage transitions are conserved across epithelial malignancies, but exhibit cancer-specific patterns. In colorectal cancer, Yang[14] identified BHLHE40 as a transcriptional driver of EMT, with malignant epithelial cells transitioning into CXCL1+ CAFs. In esophageal squamous cell carcinoma, Guo[16] reported EMT initiation in precancerous lesions, peaking before lymph node metastasis. In contrast, our study reveals a steady increase in EMT activity from early-stage to late-stage EC, with no evidence of precancerous EMT initiation or pre-metastatic peaks. Such context-dependent disparities emphasize the need for cancer-specific EMT investigations—while cross-tumor comparisons reveal conserved mechanisms, direct extrapolation risks overlooking key disease-specific features [3,4].\u003c/p\u003e\n\u003cp\u003eEmerging evidence highlights EMT as a promising therapeutic target in EC. Lengrand et al. [35]demonstrated that pharmacological blockade of netrin-1 with NP137 reduces EMT-positive tumor cells and enhances chemotherapy sensitivity, and Cassier et al. [36]validated these findings in EC mouse models and a first-in-human clinical trial—confirming the translational potential of EMT inhibition. Our identification of INHBA and POSTN as core EMT drivers provides novel therapeutic targets: inhibiting these genes could intercept the EMT cascade, suppress metastasis, and potentially sensitize EC to existing therapies. For example, targeting INHBA may disrupt both EMT and immunosuppression[27,29], while POSTN inhibition could block ECM remodeling and CAF-tumor crosstalk[30,31]—strategies worthy of preclinical and clinical exploration.\u003c/p\u003e\n\u003cp\u003eThere are several limitations to this study. First, all single-cell data are derived from stage I EC samples. This limits the comprehensive assessment of the advanced EMT characteristics of metastatic tumor niches. Second, we did not validate EMT dynamics in a mouse model of EC. Third, although we confirmed the role of INHBA and POSTN in driving EMT, we did not conduct a detailed study of upstream regulatory factors and downstream influencing factors. Finally, our functional analysis relied exclusively on gene knockdown approaches, resulting in the potential effects of overexpression being unexplored.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eOur study provides a comprehensive characterization of EMT heterogeneity and dynamic progression in EC, identifies CXCL14_CAFs/PCP4_CAFs as key EMT-associated stromal subsets, uncovers a hybrid E/M intermediate state, and validates INHBA and POSTN as core EMT regulators with prognostic and therapeutic potential. These findings deepen our understanding of EC biology, fill critical gaps in current EMT research [5,11]. As EMT inhibition emerges as a viable therapeutic approach in EC[36,37], our identification of INHBA and POSTN as central drivers offers promising targets to improve outcomes for patients with advanced or recurrent disease.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003eAcknowledgments We would also like to acknowledge the TCGA and GEO databases for offering publicly available datasets that were used in our analysis.We would like to thank all the teachers in the Daping Hospital, Army Medical University for their support.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u0026nbsp;\u003c/strong\u003eAll authors contributed to the study conception and design. \u0026nbsp;Material preparation, data collection and analysis were performed by DL, BW and JH. The first draft of the manuscript was written by DL, YM and XC all authors commented on previous versions of the manuscript.\u003c/p\u003e\n\u003cp\u003eFunding\u0026nbsp;The authors declare that financial support was received for the research of this article. This study was supported by the Basic Research Project of Army Medical University (Grant No.50122-4039).\u003c/p\u003e\n\u003cp\u003eData availability The scRNA-seq datasets generated in this study are available in the GEO ( https://www.ncbi.nlm.nih.gov/geo , Accession: GSE173682 and GSE251923). Bulk RNA-seq data and clinical records were retrieved from The Cancer Genome Atlas (TCGA-UCEC project, https://portal.gdc.cancer.gov). All datasets used in our study are from previously published studies, and detailed information can be found in the Supplementary Material. Further inquiries can be directed to the corresponding authors.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict of interest\u0026nbsp;\u003c/strong\u003eThe authors declare no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthical approval\u0026nbsp;\u003c/strong\u003eThe studies involving humans were approved by Ethics Committee of Daping Hospital，Army Medical University (Approval No.202131). The studies were conducted in accordance with the local legislation and institutional requirements.\u0026nbsp;\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eCrosbie EJ, Kitson SJ, McAlpine JN, MukhopadhyayA, Powell ME, Singh N. Endometrial cancer. Lancet. 2022;399:1412\u0026ndash;28.https://doi.org/10.1016/S0140-6736(22)00323-3\u003c/li\u003e\n\u003cli\u003eCorr BR, Erickson BK, Barber EL, Fisher CM, Slomovitz B. Advances in the management of endometrial cancer. BMJ. 2025;388:e080978. https://doi.org/10.1136/bmj-2024-080978\u003c/li\u003e\n\u003cli\u003eYang J, Antin P, Berx G, Blanpain C, Brabletz T, Bronner M, et al. Guidelines and definitions for research on epithelial-mesenchymal transition. Nat Rev Mol Cell Biol. 2020;21:341\u0026ndash;52. https://doi.org/10.1038/s41580-020-0237-9\u003c/li\u003e\n\u003cli\u003eAkhmetkaliyev A, Alibrahim N, Shafiee D,Tulchinsky E. EMT/MET plasticity in cancer and Go-or-Grow decisions inquiescence: the two sides of the same coin? Mol Cancer. 2023;22:90.https://doi.org/10.1186/s12943-023-01793-z\u003c/li\u003e\n\u003cli\u003eRen F, Wang L, Wang Y, Wang J, Wang Y, Song X, et al. Single-cell transcriptome profiles the heterogeneity of tumor cells and microenvironments for different pathological endometrial cancer and identifies specific sensitive drugs. Cell Death Dis. 2024;15:571. https://doi.org/10.1038/s41419-024-06960-8\u003c/li\u003e\n\u003cli\u003eKalluri R. The biology and function of fibroblasts in cancer. Nat Rev Cancer. 2016;16:582\u0026ndash;98. https://doi.org/10.1038/nrc.2016.73\u003c/li\u003e\n\u003cli\u003eYamamoto Y, Kasashima H, Fukui Y, Tsujio G, Yashiro M, Maeda K. The heterogeneity of cancer-associated fibroblast subpopulations: Their origins, biomarkers, and roles in the tumor microenvironment. Cancer Sci. 2023;114:16\u0026ndash;24. https://doi.org/10.1111/cas.15609\u003c/li\u003e\n\u003cli\u003eCheng PSW, Zaccaria M, Biffi G. Functional heterogeneity of fibroblasts in primary tumors and metastases. Trends Cancer. 2024;S2405-8033(24)00260-7. https://doi.org/10.1016/j.trecan.2024.11.005\u003c/li\u003e\n\u003cli\u003eBiffi G, Tuveson DA. Diversity and Biology of Cancer-Associated Fibroblasts. Physiol Rev. 2021;101:147\u0026ndash;76. https://doi.org/10.1152/physrev.00048.2019\u003c/li\u003e\n\u003cli\u003eCords L, de Souza N, Bodenmiller B. Classifying cancer-associated fibroblasts-The good, the bad, and the target. Cancer Cell. 2024;42:1480\u0026ndash;5. https://doi.org/10.1016/j.ccell.2024.08.011\u003c/li\u003e\n\u003cli\u003eYu Z, Zhang J, Zhang Q, Wei S, Shi R, Zhao R, et al. Single-cell sequencing reveals the heterogeneity and intratumoral crosstalk in human endometrial cancer. Cell Proliferat. 2022;55:e13249. https://doi.org/10.1111/cpr.13249\u003c/li\u003e\n\u003cli\u003eStuart T, Satija R. Integrative single-cell analysis. Nat Rev Genet. 2019;20:257\u0026ndash;72. https://doi.org/10.1038/s41576-019-0093-7\u003c/li\u003e\n\u003cli\u003ePapalexi E, Satija R. Single-cell RNA sequencing to explore immune cell heterogeneity. Nat Rev Immunol. 2018;18:35\u0026ndash;45. https://doi.org/10.1038/nri.2017.76\u003c/li\u003e\n\u003cli\u003eYang S, Zhang D, Sun Q, Nie H, Zhang Y, Wang X,et al. Single-Cell and Spatial Transcriptome Profiling Identifies theTranscription Factor BHLHE40 as a Driver of EMT in Metastatic ColorectalCancer. Cancer Res. 2024;84:2202\u0026ndash;17. https://doi.org/10.1158/0008-5472.CAN-23-3264\u003c/li\u003e\n\u003cli\u003eWu Q, Jiang G, Sun Y, Li B. Reanalysis of single-cell data reveals macrophage subsets associated with the immunotherapy response and prognosis of patients with endometrial cancer. Exp Cell Res. 2023;430:113736. https://doi.org/10.1016/j.yexcr.2023.113736\u003c/li\u003e\n\u003cli\u003eGuo D, Sheng K, Zhang Q, Li P, Sun H, Wang Y, et al. Single-cell transcriptomic analysis reveals the landscape of epithelial-mesenchymal transition molecular heterogeneity in esophageal squamous cell carcinoma. Cancer Lett. 2024;587:216723. https://doi.org/10.1016/j.canlet.2024.216723\u003c/li\u003e\n\u003cli\u003eWang S, Sun S-T, Zhang X-Y, Ding H-R, Yuan Y, He J-J, et al. The Evolution of Single-Cell RNA Sequencing Technology and Application: Progress and Perspectives. Int J Mol Sci. 2023;24:2943. https://doi.org/10.3390/ijms24032943\u003c/li\u003e\n\u003cli\u003eGonz\u0026aacute;lez-Mart\u0026iacute;nez S, P\u0026eacute;rez-Mies B, Cort\u0026eacute;s J, Palacios J. Single-cell RNA sequencing in endometrial cancer: exploring the epithelial cells and the microenvironment landscape. Front Immunol. 2024;15:1425212. https://doi.org/10.3389/fimmu.2024.1425212\u003c/li\u003e\n\u003cli\u003eRen X, Liang J, Zhang Y, Jiang N, Xu Y, Qiu M, et al. Single-cell transcriptomic analysis highlights origin and pathological process of human endometrioid endometrial carcinoma. Nat Commun. 2022;13:6300. https://doi.org/10.1038/s41467-022-33982-7\u003c/li\u003e\n\u003cli\u003eHashimoto S, Tabuchi Y, Yurino H, Hirohashi Y, Deshimaru S, Asano T, et al. Comprehensive single-cell transcriptome analysis reveals heterogeneity in endometrioid adenocarcinoma tissues. Sci Rep-uk. 2017;7:14225. https://doi.org/10.1038/s41598-017-14676-3\u003c/li\u003e\n\u003cli\u003eWang J, Meng F, Mao F. Single cell sequencing analysis and transcriptome analysis constructed the liquid-liquid phase separation(LLPS)-related prognostic model for endometrial cancer. Front Oncol. 2022;12:1005472. https://doi.org/10.3389/fonc.2022.1005472\u003c/li\u003e\n\u003cli\u003eCochrane DR, Campbell KR, Greening K, Ho GC, Hopkins J, Bui M, et al. Single cell transcriptomes of normal endometrial derived organoids uncover novel cell type markers and cryptic differentiation of primary tumours. J Pathol. 2020;252:201\u0026ndash;14. https://doi.org/10.1002/path.5511\u003c/li\u003e\n\u003cli\u003eChhabra Y, Weeraratna AT. Fibroblasts in cancer: Unity in heterogeneity. Cell. 2023;186:1580\u0026ndash;609. https://doi.org/10.1016/j.cell.2023.03.016\u003c/li\u003e\n\u003cli\u003eBrabletz S, Schuhwerk H, Brabletz T, Stemmler MP. Dynamic EMT: a multi-tool for tumor progression. EMBO J. 2021;40:e108647. https://doi.org/10.15252/embj.2021108647\u003c/li\u003e\n\u003cli\u003eDeng Z, Fan T, Xiao C, Tian H, Zheng Y, Li C, et al. TGF-\u0026beta; signaling in health, disease, and therapeutics. Signal Transduct Target Ther. 2024;9:61. https://doi.org/10.1038/s41392-024-01764-w\u003c/li\u003e\n\u003cli\u003eColak S, Ten Dijke P. Targeting TGF-\u0026beta; Signaling in Cancer. Trends Cancer. 2017;3:56\u0026ndash;71. https://doi.org/10.1016/j.trecan.2016.11.008\u003c/li\u003e\n\u003cli\u003eHu Y, Recouvreux MS, Haro M, Taylan E, Taylor-Harding B, Walts AE, et al. INHBA(+) cancer-associated fibroblasts generate an immunosuppressive tumor microenvironment in ovarian cancer. NPJ Precis Oncol. 2024;8:35. https://doi.org/10.1038/s41698-024-00523-y\u003c/li\u003e\n\u003cli\u003eWang J-J, Chen D-X, Zhang Y, Xu X, Cai Y, Wei W-Q, et al. Elevated expression of the RNA-binding protein IGF2BP1 enhances the mRNA stability of INHBA to promote the invasion and migration of esophageal squamous cancer cells. Exp Hematol Oncol. 2023;12:75. https://doi.org/10.1186/s40164-023-00429-8\u003c/li\u003e\n\u003cli\u003eLi F, Gu L, Tong Y, Chen R, Chen S, Yu X, et al. INHBA promotes tumor growth and induces resistance to PD-L1 blockade by suppressing IFN-\u0026gamma; signaling. Acta Pharmacol Sin. Nature Publishing Group; 2025;46:448\u0026ndash;61. https://doi.org/10.1038/s41401-024-01381-x\u003c/li\u003e\n\u003cli\u003eWang H, Liang Y, Liu Z, Zhang R, Chao J, Wang M, et al. POSTN+ cancer-associated fibroblasts determine the efficacy of immunotherapy in hepatocellular carcinoma. J Immunother Cancer. 2024;12:e008721. https://doi.org/10.1136/jitc-2023-008721\u003c/li\u003e\n\u003cli\u003eLin S-C, Liao Y-C, Chen P-M, Yang Y-Y, Wang Y-H, Tung S-L, et al. Periostin promotes ovarian cancer metastasis by enhancing M2 macrophages and cancer-associated fibroblasts via integrin-mediated NF-\u0026kappa;B and TGF-\u0026beta;2 signaling. J Biomed Sci. 2022;29:109. https://doi.org/10.1186/s12929-022-00888-x\u003c/li\u003e\n\u003cli\u003eZheng Q-M, Lu J-J, Zhao J, Wei X, Wang L, Liu P-S. Periostin Facilitates the Epithelial-Mesenchymal Transition of Endometrial Epithelial Cells through ILK-Akt Signaling Pathway. Biomed Res Int. 2016;2016:9842619. https://doi.org/10.1155/2016/9842619\u003c/li\u003e\n\u003cli\u003eDe Oliveira Macena Y, Cezar MEN, Lira CBF, De Oliveira LBDM, Almeida TN, Costa ADAV, et al. The roles of periostin derived from cancer-associated fibroblasts in tumor progression and treatment response. Cancer Metastasis Rev. 2024;44:11. https://doi.org/10.1007/s10555-024-10233-3\u003c/li\u003e\n\u003cli\u003eNull JL, Kim DJ, McCann JV, Pramoonjago P, Fox JW, Zeng J, et al. Periostin+ Stromal Cells Guide Lymphovascular Invasion by Cancer Cells. Cancer Res. 2023;83:2105\u0026ndash;22. https://doi.org/10.1158/0008-5472.CAN-22-2412\u003c/li\u003e\n\u003cli\u003eLengrand J, Pastushenko I, Vanuytven S, Song Y, Venet D, Sarate RM, et al. Pharmacological targeting of netrin-1 inhibits EMT in cancer. Nature. 2023;620:402\u0026ndash;8. https://doi.org/10.1038/s41586-023-06372-2\u003c/li\u003e\n\u003cli\u003eCassier PA, Navaridas R, Bellina M, Rama N, Ducarouge B, Hernandez-Vargas H, et al. Netrin-1 blockade inhibits tumour growth and EMT features in endometrial cancer. Nature. 2023;620:409\u0026ndash;16. https://doi.org/10.1038/s41586-023-06367-z\u003c/li\u003e\n\u003cli\u003eXia X, Yin K, Wang S. Targeting of netrin-1 by monoclonal antibody NP137 inhibits the EMT in cancer. J Immunother Cancer. 2024;12:e008937. https://doi.org/10.1136/jitc-2024-008937\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Endometrial cancer, single cell sequencing, epithelial-mesenchymal transition, INHBA, POSTN","lastPublishedDoi":"10.21203/rs.3.rs-8267986/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8267986/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Endometrial cancer (EC) incidence rises with aging and obesity epidemics. While early-stage EC has favorable outcomes, metastatic disease portends poor survival, necessitating biomarkers for metastasis prediction. Epithelial-mesenchymal transition (EMT) drives EC progression, but its dynamics and core regulators in clinical specimens remain uncharacterized.We performed single-cell RNA sequencing on 7 EC, integrated with bulk RNA-seq from 521 TCGA-UCEC cases. Cellular trajectories were reconstructed via pseudotime analysis. A specific 47 gene signature was derived by consensus non-negative matrix factorization. Core EMT regulators were functionally validated in vitro. We mapped a sequential EMT progression axis in clinical specimens, identifying a transitional hybrid E/M state. A specific 47-gene signature stratified patients into two groups with significantly divergent survival; functional validation pinpointed INHBA and POSTN as core EMT drivers, whose knockdown suppressed migration and invasion, reversed EMT, and impaired proliferation, with stage-dependent upregulation confirmed in advanced EC. This study defines INHBA and POSTN as central EMT regulators with prognostic and therapeutic potential, providing the first map of EMT dynamics in clinical EC specimens.","manuscriptTitle":"Single-Cell analysis Uncovers the Dynamic Process of EMT in Endometrial Cancer and the Core Roles of INHBA and POSTN in Regulating Tumor Malignant Phenotypes","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-03 15:36:24","doi":"10.21203/rs.3.rs-8267986/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"689fa5b3-20ba-45b5-aa4e-905f9ef82cda","owner":[],"postedDate":"February 3rd, 2026","published":true,"recentEditorialEvents":[{"type":"decision","content":"Withdrawn","date":"2026-05-12T07:30:37+00:00","index":"","fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-05-12T07:44:07+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-03 15:36:24","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8267986","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8267986","identity":"rs-8267986","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}