Downregulation of EZR inhibits esophageal squamous cell carcinoma progression through the impairment of focal adhesion activation and the EGFR-mediated PI3K/AKT pathway

Abstract

Background

Esophageal squamous cell carcinoma (ESCC) is a prevalent gastrointestinal malignancy in China and is associated with a poor prognosis and limited therapeutic options. EZR encodes Ezrin, a key scaffold protein that bridges the cell membrane and cytoskeleton. Its aberrant upregulation and carcinogenic role in driving malignant progression across numerous human malignancies has been established. However, the expression profile, specific biological functions, and underlying molecular mechanisms of EZR in ESCC remain largely unclear, which has limited the exploration and development of novel targeted therapeutic strategies against ESCC. This study aims to explore the role of EZR in the malignant progression of ESCC, identify its potential downstream molecules, and screen for its potential interacting proteins, so as to reveal the key mechanism by which EZR regulates ESCC progression.

Methods

Bioinformatics analysis was performed using The Cancer Genome Atlas (TCGA) and Genotype-Tissue Expression (GTEx) datasets. Meanwhile, EZR protein expression in clinical ESCC tissues and paired adjacent normal tissues was determined by immunohistochemistry (IHC) and further validated in ESCC cell lines via Western blotting and quantitative real-time polymerase chain reaction (qRT-PCR). Stable EZR-knockdown ESCC cell lines were constructed, and in vitro functions (proliferation, colony formation, migration, and invasion) were evaluated via Cell Counting Kit-8 (CCK-8), colony formation, wound healing, and Transwell assays. In vivo tumor growth and metastasis were assessed with nude mouse xenograft and zebrafish tumor migration models. Enrichment analyses, including Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), and gene set enrichment analysis (GSEA), were conducted to examine regulatory pathways. Focal adhesion activity and EGFR/PI3K-AKT pathway activation were detected via Western blotting. Exogenous coimmunoprecipitation (Co-IP) with flag antibody in HEK293T cells was performed to verify the Ezrin-EGFR protein interaction.

Results

EZR was significantly upregulated in ESCC tissues and cell lines, and its high expression was associated with poor immune infiltration and unfavorable prognosis. EZR knockdown markedly suppressed ESCC cell proliferation, colony formation, migration, and invasion in vitro, and it inhibited xenograft growth and tumor cell distant metastasis in vivo. Regarding mechanism, EZR silencing did not alter classic epithelial-mesenchymal transition (EMT) markers but did impair focal adhesion activity to inhibit cell migration. Furthermore, EZR knockdown blocked EGF-induced EGFR phosphorylation and inactivated the downstream PI3K/AKT pathway. Exogenous Co-IP confirmed a protein-protein interaction between flag-tagged Ezrin and EGFR.

Conclusions

EZR is an oncogene that mediates the malignant progression of ESCC through two pathways: it impairs cell migration and invasion by attenuating focal adhesion activity, and inhibits cell proliferation by reducing EGFR activation-mediated PI3K-AKT signaling axis activation. These findings clarify the mechanism by which EZR contributes to ESCC progression and suggest a promising therapeutic target for ESCC treatment.

Highlight box.

Key findings

• EZR is highly expressed in esophageal squamous cell carcinoma (ESCC), and its high expression is closely correlated with the malignant progression of ESCC. EZR knockdown impairs focal adhesion activity to inhibit cell motility and blocks EGFR phosphorylation-mediated PI3K/AKT signaling pathway activation to attenuate cell proliferation; moreover, EZR can directly interact with EGFR at the protein level to regulate the activation of downstream oncogenic signaling.

What is known and what is new?

• Ezrin, encoded by EZR, is a vital membrane-cytoskeleton scaffold protein, and its aberrant expression drives tumor proliferation and metastasis in multiple malignancies, yet the regulatory mechanism of EZR in ESCC remains largely unclarified.

• This study clarified the molecular mechanism underlying ESCC proliferation and metastasis through the following findings: EZR exerts a carcinogenic effect in ESCC, EZR promotes ESCC progression by regulating focal adhesion function and the EGFR/PI3K-AKT signaling axis, and the proteins EZR and EGFR interact with one another.

What is the implication, and what should change now?

• EZR can serve as a novel prognostic biomarker and potential therapeutic target for patients with ESCC, providing a theoretical basis for the targeted therapy of ESCC. Clinically, targeted drugs or gene therapy strategies targeting the EZR-EGFR signaling axis warrant further preclinical research and development in order to overcome the current limitations in the targeted treatment of ESCC.

Introduction

According to GLOBOCAN 2022, esophageal squamous cell carcinoma (ESCC) is highly endemic in certain regions, with distinct lifestyle and environmental risk factors contributing to its pathogenesis (1).

Genetically, ESCC exhibits high interindividual and geographic heterogeneity, and no universally accepted driver genes have been identified thus far (2-6). The lack of consensus regarding its core molecular mechanisms has severely limited the development of effective targeted therapies, highlighting the urgent need to identify novel functional molecules and prognostic biomarkers to advance the precision treatment of ESCC.

EZR encodes Ezrin, a key membrane-cytoskeleton scaffolding protein involved in cell adhesion, signal transduction, and cytoskeletal remodeling (7). Under physiological conditions, Ezrin acts as a pivotal bridge connecting the plasma membrane and actin cytoskeleton. It binds membrane components via its N-terminal FERM domain and anchors to the cytoskeleton through its C-terminal C-ERMAD domain, maintaining cell integrity and mediating transmembrane signaling. Ezrin regulates morphogenesis, endocytosis, and membrane remodeling by modulating actin dynamics and calcium signaling. Ezrin deficiency leads to neonatal lethality in mice, indicating its essential and non-redundant role in development and epithelial homeostasis. Its activity is tightly controlled by post-translational modifications to ensure balanced membrane–cytoskeleton interactions (8,9). Its dysregulation is closely linked to tumorigenesis and malignant progression across multiple cancers, in which Ezrin modulates cell proliferation, invasion, and metastasis through the regulation of focal adhesion dynamics and membrane receptor signaling (10-12). However, the expression pattern, biological functions, and precise molecular mechanisms of EZR in ESCC remain poorly understood and remain to be systematically investigated.

In this study, we first analyzed EZR expression in esophageal cancer (ESCA) using The Cancer Genome Atlas (TCGA) and Genotype-Tissue Expression (GTEx) databases and further confirmed EZR upregulation in clinical ESCC tissues and cell lines via immunohistochemistry (IHC), quantitative real-time polymerase chain reaction (qRT-PCR), and Western blotting. Functional assays indicated that EZR knockdown markedly suppressed ESCC cell proliferation, colony formation, migration, and invasion in vitro and xenograft tumor growth in nude mice and distant cell migration in zebrafish in vivo. In terms of mechanism, silencing EZR reduced focal adhesion phosphorylation to impair focal adhesion activity and inhibit migration, while the disruption of direct Ezrin-EGFR interaction blocked EGF-induced EGFR phosphorylation and downstream PI3K/AKT pathway activation to suppress proliferation. Our findings clarified the oncogenic role of EZR and its dual regulatory mechanism in ESCC progression, providing a potential therapeutic target for ESCC management. We present this article in accordance with the ARRIVE and MDAR reporting checklists (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-1037/rc).

Methods

Data collection

A range of open-source, publicly accessible databases were used in this study. We accessed the RNA-sequencing data from patients with ESCA in the TCGA and the University of California, Santa Cruz (UCSC) databases. In addition, we performed partial pan-cancer analyses using SangerBox (http://www.sangerbox.com) (13) and the Xiantao Academic Public Analysis Platform (https://www.xiantaozi.com/). This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

Cell culture and treatment

The human esophageal epithelial cell line HET-1A was purchased from EallBio Biomedical Technology Co., Ltd. (Beijing, China). Human ESCC cell lines (KYSE30, KYSE150, and TE1) were obtained from Sun Yat-sen University Cancer Center, Guangdong Esophageal Cancer Institute (Guangzhou, China). Human embryonic kidney 293T cells were obtained from Zhongshan School of Medicine, Sun Yat-sen University (Guangzhou, China). Stable cell lines (KYSE30-NC, KYSE30-shEZR, KYSE150-NC, KYSE150-shEZR, and other corresponding cell lines) were established in-house by the authors.

KYSE30, KYSE150, TE1, and 293T cells and all stably transfected cell lines were cultured in high-glucose Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin solution. HET-1A cells were maintained in a specialized culture medium purchased from EallBio Biomedical Technology Co., Ltd. All cells were incubated in a humidified cell culture incubator at 37 ℃ with 5% CO₂.

RNA isolation and qRT-PCR

Total RNA from cells was extracted with TRIzol (Invitrogen, Thermo Fisher Scientific, Waltham, MA, USA) and reverse-transcribed into complementary DNA (cDNA) with the PrimeScript RT reagent kit with gDNA eraser (Takara Bio, Kusatsu, Japan). TB Green Premix Ex Taq (Takara Bio) was used to perform qRT-PCR on a LightCycler 480-II (Roche, Basel, Switzerland). We applied the 2^−ΔΔCT method and normalized to glyceraldehyde 3-phosphate dehydrogenase (GAPDH) to determine gene expression.

Western blot analysis

The cells were subjected to lysis with a lysis buffer and then subjected to centrifugation at 4 ℃ to isolate and retrieve the proteins present in the supernatant fraction. Protein concentration was determined with a bicinchoninic acid (BCA) protein assay kit (Pierce, Thermo Fisher Scientific) according to the manufacturer’s instructions. For Western blot analysis, equal amounts of protein extract were separated via sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and electrotransferred to polyvinylidene difluoride (PVDF) membranes (Pall Corp., Port Washington, NY, USA). The PVDF membranes were blocked with 5% nonfat dry milk in Tris-buffered saline with Tween-20 for 1 hour at room temperature; notably, membranes targeting phosphorylated proteins were blocked with 5% bovine serum albumin (BSA; cat. No. GC305010-100g; Servicebio, Wuhan, China) in Tris-buffered saline with Tween-20 for 1 hour, and the BSA was used for blocking to ensure the specific and stable binding of phosphorylation antibodies. Subsequently, the membranes were incubated overnight at 4 ℃ with diluted primary antibodies according to the manufacturer’s instructions. The following primary antibodies were purchased and applied in this study: anti-Ezrin [cat. No. 3145; Cell Signaling Technology (CST), Danvers, MA, USA], β-actin mouse antibody (cat. No. 66009-1-Ig; Proteintech, Rosemont, IL, USA), GAPDH (cat. No. 60004-1-Ig; Proteintech), vimentin (D21H3) XP rabbit monoclonal antibody (cat. No. 5741; CST), E-cadherin (24E10) rabbit monoclonal antibody (cat. No. 3195; CST), N-cadherin (D4R1H) rabbit monoclonal antibody (cat. No. 13116; CST), AKT serine/threonine kinase antibody (cat. No. 9272; CST), phospho-AKT (Ser473) (D9E) XP rabbit monoclonal antibody (cat. No. 4060; CST), EGFR antibody (cat. No. 2232; CST), phospho-epidermal growth factor receptor (Tyr1068) antibody (cat. No. 2234; CST), EGFR (6H11) mouse monoclonal antibody (cat. No. 201012-6H11; Zen BioScience, Beijing, China), phospho-phosphoinositide 3-kinase p85 alpha antibody (cat. No. CY6427; Abways, Shanghai, China), phosphoinositide 3-kinase p85 alpha antibody (cat. No. CY5355; Abways), phospho-paxillin (Tyr118) rabbit monoclonal antibody (cat. No. R381444; Zen BioScience), paxillin rabbit monoclonal antibody (cat. No. R25286; Zen BioScience), and vinculin rabbit monoclonal antibody (cat. No. R26085; Zen BioScience). After incubation, the membranes were washed with Tris-buffered saline with Tween-20 for several times and then incubated with the corresponding secondary antibodies for 1 hour at room temperature, including anti-rabbit immunoglobulin G (IgG) horseradish peroxidase (HRP)-linked antibody (cat. No. 7074; CST) and anti-mouse IgG HRP-linked antibody (cat. No. 7076; CST). Following secondary antibody incubation and thorough washing, immunoblots were visualized with enhanced chemiluminescence reagent (cat. No. WBKLS0500; MilliporeSigma, Burlington, MA, USA).

Immunohistochemical staining

Mouse/rabbit polymer test system universal kit (Beijing Zhongshan Golden Bridge Biotechnology Co., Ltd., Beijing, China) was used to perform IHC according to the manufacturer’s instructions. Briefly, the paraffin sections from our hospital were placed in a dry oven at 60 ℃ for 2 hours and then deparaffinized in xylene and rehydrated in graded ethanol. The sections were placed in ethylene diamine tetraacetic acid solution (pH 9.5) and underwent high-pressure repair for 10 minutes. After cooling down naturally, the sections were incubated in 3 % H₂O₂ for 10 minutes. Subsequently, the sections were blocked with goat serum for 30 minutes and incubated with a specific primary antibody at 4 ℃ overnight. Next, the sections were incubated with an enzyme-labeled anti-goat IgG polymer, stained with diaminobenzidine, and counterstained with hematoxylin. The immunoreactive score was determined by multiplying the staining intensity score by the positive cell proportion score. Quantitative statistical analysis was performed based on the scoring criteria as follows: the positive cell proportion score was graded from 0 to 4, with 0 representing no positive cells, 1 representing 1–10% positive cells, 2 representing 11–50% positive cells, 3 representing 51–80% positive cells, and 4 representing more than 80% positive cells; the staining intensity score was graded from 0 to 3, with 0 indicating no staining, 1 indicating weak positive staining, 2 indicating moderate positive staining, and 3 indicating strong positive staining.

Plasmid construction and lentiviral transfection

shEZR sequences (including shEZR-2, shEZR-3, and shEZR-4) were synthesized by Tsingke Biotechnology Co., Ltd. (Beijing, China). Lentivirus was packaged with the three-plasmid system consisting of psPAX2, VSV-G, and pLVX-Puro-shEZR. The EZR-FLAG overexpression plasmid and EGFR overexpression plasmid were purchased from Miaoling Biotechnology Co., Ltd. (Wuhan, China). Lentiviral infection and plasmid transfection were performed according to our previously reported procedures (14).

Cell Counting Kit-8 (CCK-8) analysis

ESCC cells were seeded into 96-well plates and cultured for 24, 48, and 72 hours, respectively. Subsequently, 10 µL of CCK-8 solution (cat. No. C6005; New Cell & Molecular Biotech, Suzhou, China) was added to each well and incubated for 3 hours, and the optical density value at 450 nm was measured with the Sunrise microplate reader (Tecan Group Ltd., Männedorf, Switzerland).

Colony formation assay

A total of 800 viable ESCC cells were seeded into six-well culture plates and continuously cultured for 14 days to facilitate colony formation. Following incubation, the resulting cell colonies were gently rinsed with phosphate-buffered saline (PBS), fixed thoroughly, and then stained with 1% crystal violet solution for 30 minutes.

Wound healing assay

Cell migration capacity was evaluated via the wound healing scratch assay. Briefly, trypsin-digested ESCC cells were inoculated into culture plates equipped with scratch inserts (cat. No. CR211-2; Coolrun Life Science Technology Co., Ltd., Shenzhen, China). After full cell adherence, the inserts were gently removed to generate a uniform linear wound. The cells were then cultured in fresh serum-free medium, and images were captured at preset time points to compare wound closure dynamics.

Transwell migration and invasion assays

Transwell chambers (cat. No. 353097; Corning Inc., Corning, NY, USA) were used to assess cell migration and invasion abilities. For the invasion assay, Matrigel (cat. no. 356234; Corning Inc.) was diluted with serum-free culture medium, coated onto the upper surface of the bottom membrane in the upper chamber, and incubated at 37 ℃ for 3 hours to allow for complete gel polymerization; this step was omitted for the migration assay. The lower chambers were filled with complete culture medium, and 5×10⁴ viable ESCC cells were seeded into the upper chambers, which was followed by incubation at 37 ℃ for 24 hours to facilitate transmembrane movement. Cells that migrated to the lower surface of the membrane were then fixed and stained for subsequent quantification.

Immunofluorescence

The cells were seeded in culture dish with glass bottom (Biosharp Life Sciences, Hefei, China). After fixed with 4% paraformaldehyde and permeabilized with 0.25% Triton X-100 (Biosharp Life Sciences), cells were blocked with goat serum at 25 ℃ for 30 minutes and incubated with primary antibodies at 4 ℃ overnight. The cells were then incubated with fluorescent-labeled IgG polymer at 25 ℃ for 1 hour. Nuclei were stained with 4',6-diamidino-2-phenylindole (DAPI) (Servicebio, Wuhan, China). Images were obtained with an LSM 880 confocal microscope (Zeiss, Wetzlar, Germany).

Coimmunoprecipitation (Co-IP) assay

After cotransfection with EZR-FLAG overexpression plasmid and EGFR overexpression plasmid, HEK-293T cells were cultured in accordance with the standard cell culture protocol. Subsequently, protein extraction and Co-IP were performed with a Co-IP kit (cat. No. HY-K0202K; MedChem Express, Monmouth Junction, NJ, USA) according to the manufacturer’s instructions, and the target proteins were finally detected by Western blotting.

Functional enrichment analysis

Transcriptome sequencing, as well as Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses, was conducted by MajorBio (Shanghai, China). Gene set enrichment analysis (GSEA) was conducted with the R “clusterProfiler” package (The R Foundation of Statistical Computing, Vienna, Austria). The relevant gene sets were downloaded from the Molecular Signatures Database (http://www.gsea-msigdb.org/gsea/downloads.jsp) (15). P<0.05 was considered statistically significant in the GO and KEGG analyses. Based on the median expression of EZR, we divided the patients into high and low groups. We performed GSEA based on the GO gene sets. A P value <0.05 and a false-discovery rate <0.25 were considered to indicate statistical significance in the GSEA.

In vivo xenograft tumor model

A total of 16 4-week-old male BALB/c-nude mice were purchased from the Experimental Animal Center (Production Zone, East Campus) of Sun Yat-sen University [certificate No. SCXK (Yue) 2021-0029]. For the subcutaneous xenograft assay, KYSE30-NC and KYSE30-shEZR cells in the logarithmic growth phase were harvested, with culture medium refreshed 12 hours prior to collection. Cells were digested with trypsin and resuspended in serum-free medium and then blended with Matrigel at a 1:1 ratio on ice to a final concentration of 2.5×10⁷ cells/mL. Each nude mouse was subcutaneously injected with 200 µL of the prepared cell mixture. Mice were euthanized once the tumor volume reached 1,500 mm³, and subcutaneous xenograft tumors were dissected for imaging and subsequent analysis. All animal procedures were approved by the Institutional Animal Care and Use Committee (IACUC) of North Campus, Sun Yat-sen University (approval No. SYSU-IACUC-2025-002849), in compliance with national guidelines for the care and use of animals.

Zebrafish cell migration assay

ESCC cells were trypsinized, centrifuged at 1,000 rpm for 4 minutes, and resuspended in complete medium for counting. Cells were labeled with 10 µM of DiI (cat. No. C1991S; Beyotime Biotechnology, Shanghai, China) in a 37 ℃ water bath for 30 minutes under dark conditions and then centrifuged and resuspended to a final concentration of 2×10⁷ cells/mL. Tumor cell microinjection into the yolk sac was performed with a stereomicroscopic injection system. Briefly, 150–200 labeled cells were injected into 2- to 3-day-old zebrafish embryos, which was followed by 7 days of incubation for subsequent observation. Each group included 10 zebrafish.

Statistical analysis

Experiments were repeated three times, and data are presented as the mean ± standard deviation. Data were analyzed with SPSS version 26.0 (IBM Corp., Armonk, NY, USA) and GraphPad Prism version 10.0 (Dotmatics, Boston, MA, USA). One-way analysis of variance followed by the Tukey post-hoc test was used to analyze differences between multiple groups, and the Student’s t-test was performed to assess the differences between the two groups. P<0.05 was considered to indicate a statistically significant difference.

Results

EZR acts as an oncogene in multiple human malignancies and exhibits aberrant overexpression in ESCC tissues and cell lines

To systematically evaluate EZR expression across different tumor types, we analyzed its transcriptional levels in the TCGA and GTEx databases (Figure 1A), which revealed that EZR was highly expressed in 24 distinct tumor types. Receiver operating characteristic (ROC) curve analysis was further performed to assess the diagnostic potential of EZR expression (Figure 1B), demonstrating high diagnostic accuracy in digestive tract malignancies. Additionally, we calculated immune scores to determine the correlation between EZR expression and tumor immune infiltration (Figure 1C). The results indicated a significant negative correlation between EZR expression and immune cell infiltration degree in six malignant tumors, including ESCA, which is indicative of a poor prognostic phenotype. Subsequently, Western blotting and qRT-PCR assays were employed to detect EZR expression in human ESCC cell lines and normal human esophageal epithelial cells (Figure 1D,1E). Moreover, IHC staining was performed on pathological sections from 44 patients who underwent radical esophagectomy at the Sixth Affiliated Hospital, Sun Yat-sen University between 2020 and 2021 (Figure 1F). Collectively, these results confirmed that EZR was highly expressed in both ESCC cell lines and clinical tissues.

Figure 1.

Pan-cancer analysis of EZR expression and its clinical and biological significance in ESCC. (A) Pan-cancer analysis of EZR transcriptional levels in 33 tumor types via integrated TCGA and GTEx datasets, showing EZR upregulation in 24 different malignancies. (B) ROC curve of EZR expression in digestive tract tumors, demonstrating high diagnostic accuracy for malignant transformation. (C) Immune score analysis revealing a significant negative correlation between EZR expression and immune cell infiltration in six malignant tumors (ESCA included), indicating a poor prognostic phenotype. (D,E) Validation of EZR expression in ESCC cell lines (KYSE30, KYSE150, and TE1) and normal esophageal epithelial cells (HET-1A) via (D) Western blotting and (E) qRT-PCR. (F) Representative IHC images showing EZR expression in ESCC tissues and paired adjacent normal tissues from 44 clinical samples (IHC staining; scale bar =500 µm). The full name of the TCGA abbreviations sees the website: https://gdc.cancer.gov/resources-tcga-users/tcga-code-tables/tcga-study-abbreviations. **, P<0.01; ****, P<0.0001; -, not significant. AUC, area under the curve; CI, confidence interval; COAD, colon adenocarcinoma; ESCA, esophageal cancer; ESCC, esophageal squamous cell carcinoma; FPR, false positive rate; GTEx, Genotype-Tissue Expression; IHC, immunohistochemistry; IRS, immunoreactive score; PAAD, pancreatic adenocarcinoma; qRT-PCR, quantitative real-time polymerase chain reaction; READ, rectum adenocarcinoma; ROC, receiver operating characteristic; STAD, stomach adenocarcinoma; TCGA, The Cancer Genome Atlas; TPR, true positive rate.

EZR knockdown markedly suppressed the proliferation, migration, and invasion of ESCC cells in vitro

Stable EZR-knockdown cell lines (KYSE30-shEZR and KYSE150-shEZR) and corresponding negative control cells (KYSE30-NC and KYSE150-NC) were constructed via lentiviral transduction in KYSE30 and KYSE150 ESCC cells. The knockdown efficiency at both the messenger RNA (mRNA) and protein levels was validated via qRT-PCR and Western blot assays, respectively (Figure 2A,2B). Further functional assessments demonstrated that EZR knockdown notably suppressed the long-term clonogenic capacity of ESCC cells (Figure 2C) and weakened short-term proliferative activity (Figure 2D). Additionally, both wound healing (Figure 2E) and Transwell (Figure 2F) assays verified that EZR knockdown significantly impaired the migratory and invasive phenotypes of ESCC cells.

Figure 2.

EZR knockdown inhibited the proliferation, migration, and invasion of ESCC cells in vitro. Stable EZR knockdown (KYSE30-shEZR and KYSE150-shEZR) and negative control (KYSE30-NC and KYSE150-NC) cell lines were established via lentiviral transduction. (A,B) The knockdown efficiency of EZR was verified at the mRNA level via qPCR and at the protein level via Western blotting. shEZR-2, shEZR-3, and shEZR-4 represent three distinct EZR knockdown sequences. The knockdown efficiency of these sequences was first verified in KYSE30 cells, and the shEZR-2 sequence was then selected to establish stable EZR knockdown in KYSE150 cells. All EZR-knockdown cells used in subsequent experiments were stable cell lines constructed with the shEZR-2 sequence. (C) Long-term clonogenic capacity of ESCC cells was assessed via colony formation assay (crystal violet staining). (D) The short-term proliferative activity of ESCC cells after EZR knockdown was detected via CCK-8 assay. (E) Wound-healing assay was performed to evaluate the migratory ability of ESCC cells. (F) Transwell assay was conducted to determine the invasive ability of ESCC cells (crystal violet staining; original magnification ×100). Data are presented as the mean ± SD of three independent experiments. *, P<0.05; **, P<0.01; ***, P<0.001; ****, P<0.0001 vs. the NC group. CCK-8, Cell Counting Kit-8; ESCC, esophageal squamous cell carcinoma; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; mRNA, messenger RNA; NC, negative control; qPCR, quantitative polymerase chain reaction; SD, standard deviation.

EZR knockdown reduced the migration ability of ESCC cells by attenuating focal adhesion activity

To clarify the mechanism underlying the impaired migratory capacity of ESCC cells induced by EZR knockdown, we first performed Western blot analysis to detect the classical epithelial-mesenchymal transition (EMT) pathway, which is widely recognized as a key regulator of tumor cell migration (Figure 3A). However, no significant differences were observed in the expression levels of E-cadherin, N-cadherin, or vimentin, the core markers of the canonical EMT pathway. To further examine the potential causes underlying the inhibited migration of ESCC cells following EZR knockdown, transcriptome sequencing was carried out in KYSE30-NC and KYSE30-shEZR cell lines. GO enrichment analysis revealed that cell junction-related terms were significantly enriched (Figure 3B); this pathway involves the connection structures between cells or between cells and the extracellular matrix, as well as relevant regulatory molecules, and is closely associated with cell migration ability. Additionally, KEGG enrichment analysis demonstrated that the focal adhesion pathway was significantly enriched (Figure 3C). As a vital subcellular structure, focal adhesion participates extensively in critical biological behaviors, including cell adhesion and migration. To verify whether focal adhesion function was affected, Western blotting was conducted to examine the expression of focal adhesion-associated proteins (Figure 3D). The results indicated that the expression of vinculin, paxillin, and their phosphorylated forms—the core proteins of focal adhesions—was significantly downregulated.

Figure 3.

EZR knockdown suppressed ESCC cell migration by inactivating focal adhesion signaling. (A) Western blot analysis of core protein markers in the classical EMT pathway, including E-cadherin, N-cadherin, and vimentin, in NC and EZR-knockdown ESCC cells. (B) GO enrichment analysis of differentially expressed genes identified by transcriptome sequencing in KYSE30-NC and KYSE30-shEZR cells. (C) KEGG enrichment analysis of differential genes to screen key signaling pathways correlated with cell migration. (D) Western blot detection of the expression levels of total and phosphorylated focal adhesion core proteins, including vinculin and paxillin, in ESCC cells after EZR knockdown. ESCC, esophageal squamous cell carcinoma; EMT, epithelial-mesenchymal transition; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes; NC, negative control.

Knockdown of EZR may suppress the proliferative capacity of ESCC cells by attenuating EGFR phosphorylation-mediated activation of the PI3K/AKT signaling pathway

The PI3K/AKT signaling pathway was also significantly enriched in the aforementioned transcriptome sequencing analysis (Figure 3C). As one of the most critical signaling cascades in tumor cells, the PI3K/AKT pathway is critically involved in regulating tumor cell proliferation and survival. To validate whether EZR knockdown impaired the activation of the PI3K/AKT pathway at the protein level, Western blot assays were performed for further verification (Figure 4A). Additionally, GSEA based on the TCGA database was conducted to analyze the key downstream pathways of EZR (Figure 4B). In line with our previous experimental findings and the increased activation of the ERBB pathway, we further identified significant activation of the EGFR (ERBB1) pathway. Subsequent Western blot assays revealed that the phosphorylation level of EGFR was significantly reduced following EZR knockdown (Figure 4C). To produce more distinct phenotypic changes, ESCC cells were treated with exogenous EGF stimulation, and it was found that both EGFR phosphorylation and the activation of its downstream PI3K/AKT signaling pathway were significantly reduced (Figure 4D).

Figure 4.

EZR knockdown suppressed EGFR/PI3K-AKT signaling pathway activity. (A) Western blot analysis of total and phosphorylated protein levels in the PI3K/AKT signaling pathway in NC and EZR-silenced ESCC cells. (B) GSEA of EZR-related downstream signaling pathways based on the TCGA database. (C) Western blot detection of EGFR phosphorylation levels in ESCC cells after EZR knockdown. (D) Western blot analysis of EGFR and PI3K-AKT pathway activation in ESCC cells pretreated with exogenous EGF. (E) Exogenous Co-IP assay verified the protein correlation between EZR and EGFR in HEK-293T cells. (F) Immunofluorescence colocalization assay detected the colocalization of EZR and EGFR in ESCC cells (C Plan-Apochromat 63×/1.4 oil objective). Data were obtained from three independent repeated tests, and the results are presented as the mean ± SD. ns, not significant; *, P<0.05; **, P<0.01 vs. the NC group. Co-IP, coimmunoprecipitation; Con, control; DAPI, 4',6-diamidino-2-phenylindole; ESCC, esophageal squamous cell carcinoma; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; GSEA, gene set enrichment analysis; IgG, immunoglobulin G; IP, immunoprecipitation; NC, negative control; OE, overexpression; SD, standard deviation; TCGA, The Cancer Genome Atlas.

Furthermore, we examined the underlying mechanism by which EZR protein modulates EGFR activation. For exogenous Co-IP assays, HEK-293T cells were cotransfected with FLAG-EZR overexpression plasmid and EGFR overexpression plasmid (Figure 4E), and the results revealed an association between the EZR and EGFR proteins. Subsequently, immunofluorescence colocalization assays were performed in ESCC cells (Figure 4F), which further verified the colocalization of Ezrin and EGFR, suggesting a potential protein-level correlation between the two proteins. Collectively, these data demonstrated that EZR knockdown could dampen the phosphorylation and activation of EGFR, subsequently inactivating the downstream PI3K/AKT signaling cascade and ultimately inhibiting the proliferative phenotype of ESCC cells.

The zebrafish tumor migration model and the xenograft mouse tumor formation model were used to verify the effect of EZR knockdown on the migration and proliferation ability of ESCC cells, respectively, in vivo

To determine whether the regulatory effects of EZR knockdown on the migratory and proliferative abilities of ESCC cells observed in vitro could be recapitulated in vivo, we established a zebrafish tumor cell migration model and a BALB/c nude mouse subcutaneous xenograft model, respectively. The number of tumor cells migrating from the yolk sac to the zebrafish tail was notably decreased in the EZR-knockdown group as compared with the corresponding negative control groups (Figure 5A,5B). In the nude mouse xenograft assay, tumors formed by EZR-knockdown ESCC cells exhibited markedly reduced tumor volume and weight relative to the control treatment under identical housing conditions and culture duration (Figure 5C-5E).

Figure 5.

EZR knockdown inhibited ESCC cell proliferation and migration in vivo. (A,B) Representative images (A) and quantitative analysis (B) of zebrafish xenograft models for assessing ESCC cell migration after EZR knockdown, with migrated cell numbers being counted. (C-E) Images of xenograft tumors (C), tumor weight analysis (D), and tumor volume growth curves (E) in BALB/c nude mice injected with NC and EZR-knockdown ESCC cells. Data are presented as the mean ± SD of three independent replicates. ***, P<0.001; ****, P<0.0001 vs. the NC group. ESCC, esophageal squamous cell carcinoma; NC, negative control; SD, standard deviation.

Overall, the results of the in vivo animal experiments were highly consistent with our in vitro functional findings. Taken together, these data confirm that EZR knockdown can significantly suppress the proliferation and migration of ESCC cells in vivo.

Discussion

ESCC is a common digestive tract malignancy, characterized by difficult early diagnosis, severe late-stage metastasis, poor prognosis, and pronounced therapy resistance. The initiation and progression of ESCC is a complex pathological process involving multiple genes and signaling pathways. Thus, identifying critical driver genes and clarifying their molecular regulatory mechanisms may provide critical clinical insights into novel ESCC therapeutic targets (16). In this study, we systematically analyzed the biological functions and core molecular mechanisms of EZR in the pathogenesis of ESCC by integrating bioinformatics analysis, clinical tissue validation, in vitro cell functional assays, in vivo nude mouse xenograft and zebrafish tumor migration models, and molecular mechanism experiments. We confirmed that EZR exerts oncogenic effects through two regulatory pathways. It modulates focal adhesion function to regulate cell migration and activates the downstream PI3K/AKT pathway via Ezrin-EGFR functional crosstalk to drive malignant cell proliferation. These findings may help form the experimental and theoretical foundations for targeted ESCC therapy.

Our previous study demonstrated that FOXF2, a novel tumor-suppressive transcription factor, mediates aberrant EZR expression upon inactivation, thereby promoting multiple malignant biological behaviors of ESCC (14). Accordingly, we hypothesized that EZR acts directly as an oncogene in cancer development. Other research indicates that Ezrin, encoded by EZR, is a key membrane-cytoskeleton scaffold protein, which is abnormally overexpressed in several epithelial malignancies and promotes tumor progression through the regulation of EMT and other processes (17-21). However, the expression profile, clinical correlation, and specific molecular mechanisms of EZR in ESCC remain poorly elucidated. To verify our bioinformatics-based predictions, we collected clinical ESCC and paired adjacent normal tissues for IHC validation, and the results were highly consistent with bioinformatics analysis. Further cell-level validation confirmed that EZR was significantly upregulated in ESCC at both the mRNA and protein levels, verifying its oncogenic role and the reliability of bioinformatics in tumor driver gene screening.

Tumor metastasis is traditionally considered to rely heavily on EMT activation, which enables epithelial tumor cells to acquire stronger migratory and invasive capabilities. However, accumulating evidence has confirmed that cancer cells can initiate metastasis through diverse EMT-independent mechanisms (22,23). In this study, we found that knockdown of EZR failed to alter the expression levels of classic EMT markers in ESCC cells, suggesting that EZR-mediated metastatic promotion does not depend on canonical EMT phenotypic switching. As a typical membrane-cytoskeleton linker protein, Ezrin predominantly regulates cell surface mechanical properties and focal adhesion dynamics rather than inducing epithelial-mesenchymal transformation. Focal adhesions are core structures mediating cell-extracellular matrix adhesion, and their dynamic assembly-disassembly balance is crucial for cell migration (24,25). Attenuated focal adhesion activity directly impairs cell-matrix adhesion and cytoskeletal remodeling, thus inhibiting cell migration and invasion (26). A study has reported that ezrin regulates focal adhesion complex stability and calpain-1 localization in breast cancer, modulating focal adhesion turnover (27). However, the specific mechanisms by which ezrin maintains focal adhesion function in ESCC remain unclear. Our results preliminarily confirmed that EZR knockdown suppresses ESCC cell migration and invasion by impairing focal adhesion function and is consistent with ezrin’s inherent physiological roles in cytoskeleton regulation and signal transduction, providing a direct molecular explanation for the observed migration phenotypes. Therefore, we speculate that EZR promotes ESCC cell migration mainly by stabilizing focal adhesion assembly and sustaining downstream oncogenic signaling, enhancing tumor cell motility in an EMT-independent manner. This finding explains the biological characteristic that some ESCC cells retain intact epithelial features while still possessing strong metastatic potential.

Regarding the core mechanism of EZR-regulated ESCC cell proliferation, we focused on pathways identified by GSEA and KEGG enrichment analyses and conducted an in-depth validation of the EGFR-PI3K/AKT axis, a core finding of this study. EGFR is a key cell membrane growth factor receptor, and its aberrant activation and downstream PI3K/AKT signaling constitute the core axis driving ESCC malignant proliferation, representing a vital ESCC therapeutic target (28). Yet, the regulatory mechanisms of EGFR activation in ESCC remain incompletely clarified. We found that EZR knockdown markedly reduced EGF-induced EGFR phosphorylation and downstream PI3K/AKT pathway activation, indicating EZR is as a critical regulator of this signaling axis.

Studies on Ezrin-EGFR crosstalk are sparse and have primarily focused on indirect regulatory mechanisms in lung cancer EGFR-targeted therapy (29). A report shows that EGFR is highly expressed in ESCC despite low mutation rates, while anti-EGFR monotherapy yields limited survival benefits (30). Based on our findings, we believe that investigating combined Ezrin-EGFR targeted therapy can help improve the survival of patients with EGFR-positive ESCC.

The principal innovations of this study are as follows: first, we systematically confirmed the oncogenic role of EZR in ESCC via the integration of bioinformatics, clinical specimens, and functional experiments, with clear clinical translational value. Second, this study is the first to identify the dual regulatory mechanisms of EZR in ESCC, which include the regulation of proliferation via the EGFR-PI3K/AKT pathway and of migration/invasion regulation via focal adhesion function; no evidence of classical EMT pathway involvement was found. Third, we verified the functional protein association and colocalization of Ezrin and EGFR in ESCC, revealing a novel upstream mechanism for EGFR activation, which may serve as a therapeutic target for ESCC.

This study involved several limitations that should be addressed. First, the clinical sample size should be expanded and prognostic data integrated to improve the reliability of EZR as a prognostic biomarker of ESCC. Second, the specific functional domains mediating Ezrin-EGFR association should be further explored and targeted inhibitors developed to improve the precision of ESCC therapy. Third, immunocompetent mouse models should be established to investigate the EZR-mediated regulation of the tumor immune microenvironment. Finally, a multi-index detection system combining EZR, EGFR, p-AKT, and focal adhesion proteins should be constructed to improve the early diagnosis and prognostic evaluation of ESCC.

Conclusions

In conclusion, our findings comprehensively demonstrate that EZR is highly expressed in ESCC and exerts oncogenic functions through two mechanisms: regulating cell migration and invasion through focal adhesion modulation and driving cell proliferation via EGFR-mediated PI3K/AKT pathway activation. These results clarify the biological functions and core molecular mechanisms of EZR in ESCC, providing critical experimental and theoretical support the early diagnosis, prognostic assessment, and targeted therapy of ESCC.

Supplementary

The article’s supplementary files as

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. All animal experiments were performed under a project license (No. SYSU-IACUC-2025-002849) granted by the Institutional Animal Care and Use Committee (IACUC) of North Campus, Sun Yat-sen University, in compliance with national guidelines for the care and use of animals.

Data Sharing Statement

Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-1037/dss

jtd-18-05-530-dss.pdf ^{(95KB, pdf)}

DOI: 10.21037/jtd-2026-1037