The roles EpCAM plays to enhance the malignancy of gastric cancer

Gastric cancer (GC) remains a global challenge due to its high morbidity and mortality rates especially in Asia as well as poor response to treatment. As a member of the adhesion protein family and transmembrane glycoprotein, EpCAM expressed excessively in cancer cells including GC cells. The database assay showed that EpCAM is excessively expressed and easily mutated in cancers, especially in early stage of GC. To explore the roles EpCAM plays in oncogenesis and progression of GC, the expression of EpCAM was deleted in GC cells with CRISPR/Cas9 method, and then the changes of cell proliferation, apoptosis, motility and motility associated microstructures in EpCAM-deleted GC cells (EpCAM-/-SGC7901) were detected to evaluate the rules EpCAM played. The results showed that EpCAM deletion caused cell proliferation, motility and the development of motility-relevant microstructures inhibited significantly, apoptotic trend and contact inhibition enhanced in EpCAM-deleted GC cells. The results of western blot suggested that EpCAM modulates the expression of epithelial/endothelial mesenchymal transition (EMT) correlated genes. All results as above indicated that EpCAM plays important roles to enhance the oncogenesis, malignancy and progression as a GC enhancer. Combining our results and published data together, the interaction of EpCAM with other proteins was also discussed and concluded in the discussion. Our results support that EpCAM can be considered as a novel target for the diagnosis and therapy of GC in future.


Introduction
Gastric cancer (GC) is one of the popular malignant tumors in digestive tracts worldwide (Johnston and Beckman 2019). With high incidence and mortality, GC is not easy to be early diagnosed, so its early diagnosis is important to successful therapy and improve the prognosis of patients (Layke and Lopez 2004;Takahashi et al. 2013). Currently, the diagnosis of GC is mainly depended on gastroscopy, but the missed diagnosis is high because the non-targeting of the method. Therefore, GC patients are usually missed the best surgery chance and chemotherapy become the only choice for their clinical treatment. But the efficacy of current chemotherapeutics for GC is very poor without prognosis improved because of no or weak targeting of the drugs.
Epithelial cell adhesion molecule (EpCAM) was identified as a major antigen on human colon cancer cells and often used as the marker of cancer stem cell (CSC) for human cancer immunotherapy (Patriarca et al. 2012;Pavšič et al. 2014). As a multifunctional transmembrane glycoprotein, EpCAM is expressed on the basolateral membrane surface of many epithelial cells (Huang et al. 2018), and plays a key role in structural integrity, adhesion of epithelial tissues and their interaction with the underlying stromal layer. EpCAM may involve in the regulation of cancer cell adhesion, proliferation, migration, stem cell differentiation and epithelial mesenchymal transition (EMT) (Porcel et al. 2019;Went et al. 2004;Munz et al. 2009;Gun et al. 2010).
In the study, to explore and validate the roles EpCAM plays in GC oncogenesis and progression, we deleted EpCAM expression in GC cells using CRISPR/Cas9 technology, and analyzed the changes of cell behaviors after the EpCAM deletion. Finally, combing our results and database assay, the roles EpCAM plays and its regulating pathway is discussed and concluded.

Cell line and cell culture
The SGC-7901 cell line was purchased from ATCC (Rockville, MD, USA). The cells were cultured in RPMI1640 medium supplemented with 10%(v/v) fetal bovine serum (Gibco, Gaithersburg, MD, USA), 100U/mL penicillin, and 100 mg/mL streptomycin at 37 °C in a humidified atmosphere containing 5% CO 2 . Cells in the exponential phase of growth were used in each experiment.

The generation of EpCAM-deleted cells
The EpCAM-deleted SGC7901 cell line was generated using CRISPR/Cas9 method. The sgRNAs (5′-CAC CGC AAT GCC AGT GTA CTT CAG T-3′) were designed to target EpCAM open reading frame (ORF) at website http:// www. broad insti tute. org/ rnai/ public/ analy sis-tools/ sgrna-design and re-designated at http:// crispr. mit. edu, the pX459 plasmid was kindly obtained from Broad Institute of MIT. T7EI was used to identify the efficiency of Cas9 targeting, and puromycin was used to select the stably transfected cells. The mutated cell clones were identified by the sequencing to targeted fragment of cell clones, and the EpCAM dual alleles deleted cell clone, EpCAM-/-SGC7901 was finally confirmed by western blot.

MTT assay
MTT was used to detect cell proliferation. The cells in logarithmic growth phase were resuscitated and inoculated in 96-well plate (6 × 10 3 cell per well). In total, 20 μL MTT (5 mg/ml, Sigma, USA) were respectively added to each well at 24 h, 48 h and 72 h. After MTT incubation for 5 h at 37 °C, the supernatant was discarded and 150 μl of dimethyl sulfoxide (DMSO) was added to each well to fully dissolve the formazan crystals. The OD value at 562 nm was determined by ELISA reader (BIO-TEK, New York, USA).

Cell clone formation
Cells were inoculated in 35 mm dish at gradiently diluted cell suspensions and incubated for 2 weeks at 37 °C and 5% CO 2 incubator. After washing twice with PBS, the dish was fixed with 4% paraformaldehyde and stained with 0.1% crystalline violet, and the cell clones were counted using hand counter.

Wound healing assay
The cells with 90% confluence rate were washed three times in PBS, and scratched a straight wound on the meddle monolayer. The scratches were imaged at 0 h, 48 h, 72 h and 96 h post the wounding, and then digitized by Gene tools (Syngene). Migration Rate = (1 − At/A0) × 100%. A0: pixel at 0 h; At: pixel at different time points post wounding.

Transwell chamber assay
Cells were seeded into the top chamber with 8 μm pores (Corning Costar Incorporated, New York, USA) by 8 × 103 cells/200 μL serum-free medium, and the bottom chamber was filled with RPMI1640 medium supplemented with 10% FBS. After incubated in incubator for 24 h, the bottom chamber was fixed in 4% paraformaldehyde for 30 min, stained with 0.1% crystal violet and imaged under an inverted light microscope for cell counting.

Coomassie brilliant blue staining
The cell culture was same as 3.2. When the cell confluence reached 90%, the cells were washed with PBS and fixed with 4% paraformaldehyde for 30 min. 1% Ttiton X-100 solution was used to permeabilize the cells for 20 min. Cells were stained with 0.2% coomassie brilliant blue R-250 (Sigma) for 10 min and imaged under an inverted microscope.

Scanning electron microscope (SEM)
After cells were cultured in six-well plates containing a small coverslips for 24 h, washed cells with PBS and fixed with 2.5% glutaraldehyde for 1 h. The cells were dehydrated in gradient ethanol dilutions, and the coverslips were imaged under a scanning electron microscope (Quanta200, Philips-FEI, Netherlands).

DAPI staining
After the steps as 5.1, cells were stained with DAPI staining solution (Kaiji Bio Co, Nanjing, China) for 5 min at room temperature in dark, and imaged under an inverted microscope.

Mitoview 633 staining
To display mitochondria, the cultured cells as above were incubated in the medium containing 100 nM Mitoview 633 (US Everbright Inc., Suzhou, China) for 30 min at 37 °C, and washed three times with PBS, and finally imaged under an inverted fluorescence microscope.

Flow cytometry
For the apoptotic cell cycle arresting assay, PI staining was used following the manufacture's instruction. For the apoptotic cell membrane assay, Annexin V-FITC following the manufacturer's instructions. The both kits were purchased from 4A Bioteck, Beijing, China. The cells were analyzed by the flow cytometry from CytoFLEX S, Beckman-Coulter, USA.

Western blot
Total protein sample was prepared using RIPA lysis buffer (Roche, Indianapolis, IN). The SDS-PAGE electrophoresis was run, and the transferred NC membranes (PALL, CA, USA) were incubated with rabbit-anti human EpCAM (1:288) and GAPDH (1:2000) purchased from Santa Cruz Biotechnology Inc. CA, USA. The membranes were developed using ECL Kit (Pierce, Shanghai, China) following the manufacture's instruction and digitalized using Image J software.

Statistical analysis
Data assay was performed in triplicate, presented as the mean ± SD, and analyzed using SPSS 13.0 software or Student's t-test. Statistical significance was defined as P < 0.5, P < 0.05 and P < 0.001.

EpCAM expression is up-regulated in GC and closely correlated with poor prognosis of GC patients
The mutation of EpCAM is pronounced in carcinogenesis. Analysis of EpCAM mutation types and mutation frequencies in cancer by the cBioPortal website (https:// www. cbiop ortal. org) reveals that EpCAM mutated by additions, deletions, inversions and translocations (Fig. 1A). In GC, the mutations of EpCAM include all genomic mutations as above plus structural mutations and amplifications (Fig. 1B). The gene transcriptome expression profiles showed that EpCAM expression in GC is significantly active compared to normal tissues (Fig. 1D). Assay by the database, Tumor IMmune Estimation Resource indicated a negative correlation of EpCAM expression with immune cell infiltration in gastric intestinal cancers, especially in GC. The result based on the assay of Cox regression model showed that EpCAM expression correlates negatively with immune cell infiltration, especially with T cells (Fig. 1C). Assay with Kaplan-Meier plotter (http:// kmplot. com/ analy sis/) explored the correlation of EpCAM expression with the poor prognosis of GC patients, and EpCAM expression is up-regulated in all grades of GC (Fig. 1E, F).

EpCAM-/-SGC7901 cell line was successfully generated using CRISPR
To study EpCAM function in its expression free experimental condition, EpCAM-/-SGC7901 cell line was developed using CRISPR/Cas9 method. To check the plasmodium transfection efficiency, under the same condition of transfection, the transfection efficiency of pcDNA6.2-GFP (pX459 contains no GFP) showed that the satisfied result at 48 h post transfection as Fig. 2A. EpCAM sgRNA was designed targeting exon 2, and the EpCAM targeting Cas9 vector was constructed using pX459 plasmid (Fig. 2B, C). The EpCAM exogenous expression vector was constructed as control expression. The EpCAM-deleted clone was identified by sequencing, the result (Fig. 2D) indicated that clones C2 and IIB1 were mutated, and C2 was identified as a single allele deleted (EpCAM ± SGC7901) clone and IIB1 as a double alleles deleted (EpCAM-/-SGC7901) clone. Finally, the mutated clones were confirmed by western blot (Fig. 2E). To check the rescuing efficiency of the EpCAM exogenously expressed, EpCAM was re-expressed in wild type and EpCAM-deleted SGC-7901 cells (Fig. 2F).

EpCAM deletion significantly inhibited the malignancy of GC cells
The results of MTT showed that EpCAM deletion or inhibition caused cell growth inhibited significantly (Fig. 3A), and its rescued expression by an exogenous method caused cell growth enhanced significantly (Fig. 3B). The results of cell clone formation assay indicated a same function manner of EpCAM in its deleted or exogenous expressed GC cells. (Fig. 3C-F).
The cell motility was checked by wound healing (Fig. 4A-D) and transwell assays (Fig. 4E-H). The results showed that EpCAM deletion or rescued expression caused cell motility changed significantly, and indicated EpCAM as an enhancer of cell motility in GC cells. In addition, the results of flow cytometry suggested that EpCAM deletion caused the SGC-7901 cell cycle arrested in G1, and EpCAM expression benefits GC cell proliferation (Fig. 4I).

EpCAM deletion significantly suppressed the development of motility-relevant microstructures in GC cells
Pseudopodia and filopodia are crucial microstructures maintaining cancer cell motility and malignancy. The results of coomassie brilliant blue and scanning electron microscopy (SEM) showed that the development of pseudopodia and filopodia was significantly inhibited in EpCAM deleted and inhibited GC cells, but rescued in re-expressed GC cells with contact inhibition also restored (Fig. 5A, B). The results above indicated EpCAM as an enhancer and maintainer of cytoskeleton structure in SGC7901 cells.

EpCAM deletion caused a apoptotic tendency in GC cells
The results of MitoScence 633 staining and DAPI staining showed that EpCAM deletion caused the mitochondrial membrane potential and permeability changed, and moderately nuclear condensation (Fig. 6A). Flow cytometry results further demonstrated EpCAM as an apoptotic suppressor in GC cells (Fig. 6B).

EpCAM deletion modulated the expression of EMT relevant genes
To evaluate the rules EpCAM played in GC oncogenesis, EMT relevant gene expressions were analyzed. The results of western blot showed that a moderate enhanced expression of E-cadherin (P < 0.01), and the suppressed expressions of MMP 9 and Snail 1 (P < 0.01) were caused by EpCAM deletion (Fig. 7A, B). The results above suggested EpCAM as an EMT enhancer in GC oncogenesis. Discussion GC is a cancer with numerous gene mutations accumulated in a longtime, and the fifth most popular cancer and the third cause of cancer death globally (Cao et al. 2021). The incidence and mortality rates of GC is consistently increased in China, and due to lack specific and sensitive early diagnostic method, most of GC input-patients have developed as advanced stage with mucosal infiltration and metastasis (Li and Yu 2019;Park and Herrero 2021;Selgrad et al. 2012;Venerito et al. 2016). Consequently, the GC patients were treated with few choices for surgery or chemotherapy, and therefore, patients get very poor prognosis usually.
As a multifunctional transmembrane glycoprotein, EpCAM consists of a larger extracellular structural domain of 265 amino acids (EpEX/EX) (Yahyazadeh Mashhadi  (Gires et al. 2020). EpCAM was reported associated with signal proliferation, immune evasion by intercellular oligomerization and protein hydrolysis (Armstrong and Eck 2003;Baeuerle and Gires 2007;Keller et al. 2019), and as a key enhancer for structural integrity, adhesion of epithelial tissues and interaction with the underlying stromal layer (Gorges et al. 2012). In recent years, EpCAM was recognized as a stem cell, especially cancer stem cell (CSC) marker, and it is overexpressed on CSC and circulating tumor cells (CTCs), and therefore its detection can be used to track and isolate CTCs (Han et al. 2020;Murakami et al. 2019;Terris et al. 2010;Spizzo et al. 2011). But, the rules EpCAM played in the oncogenesis and progression of GC is still unknown.
Based on the assay of the correlation of EpCAM expression and cancer progression using different databases, to explore the functions by that EpCAM drives the oncogenesis and progression of GC under an EpCAM free exprerimental environment, the EpCAM expression was blocked using CRISPR/Cas9 method. When the EpCAM-deleted cell line (EpCAM-/-SGC7901) was successfully generated, a series of cell biological changes about cell malignancy were detected and analyzed. The results (Fig. 1) of databases assay showed that EpCAM was usually excessively expressed in multiple cancers with multiple mutation ways (addition, deletion, inversion, translocation and amplification). Assay with Kaplan-Meier plotter indicated a positive correlation of EpCAM expression with the poor prognosis of GC patients. In addition, to show the immune active cells infiltration in GC tissue, Tumor IMmune Estimation Resource database and Cox regression model assay were handled, and demonstrated a negative correlation of EpCAM expression with immune cell infiltration, especially T cells infiltration that suggests GC as a cold cancer usually. (Fig. 1C). Cold cancer usually is not beneficial to get satisfied chemotherapeutic efficacy Jérôme and Daniela 2019).
Cell proliferation, motility, motility-relevant microstructural, contact inhibition and apoptosis are important indicator to evaluate the malignancy of tumor cells. The results (Figs. 3, 4) showed that EpCAM expression enhances the GC cell proliferation and motility significantly. The results (Fig. 5) indicated that EpCAM expression remodels the GC cell cytoskeleton with the enhancement of microfilament polymerization, and pseudopods development in GC cells. EpCAM expression inhibits GC cell apoptosis moderately (Fig. 6).
The expression of EpCAM enhances EMT relevant genes expressed excessively excepting E-cadherin (Fig. 7). E-cadherin plays an adhesion role (Na et al. 2020) and benefits oncogenesis and metastasis to form a new transplanted cancer focus (Xin and Kent-Man 2014).
Assay of PIP (protein-interaction-protein) EpCAM was analyzed via String + online tool, and the results (Fig. 8A) show that EpCAM interacts with multiple proteins including CLDN-7, CDH-1, and KRT 19, which regulates cell adhesion and junction to participate EMT. FHL-2 was reported as an important protein involved in oncogenesis and progression of cancers (Elisa et al. 2015). By inhibiting the cytotoxicity mediated by T cells, PTPRC inhibits the cyto-immunity in cancer and benefits to form cold cancer tissue (Nosho et al. 2010;Tong et al. 2021).

Conclusions
In summary, as an important marker expressed excessively in CSC, EpCAM is overexpressed in cancers, and positively correlates with the poor prognosis of GC patients. Our results demonstrated that EpCAM plays important roles as an enhancer in the carcinogenesis and development of GC, and enhances the proliferation, motility and relevant microstructure remodeling in GC cells. Online assay suggested that EpCAM mediates to form cold cancer  tissue with the increased difficulty of GC chemotherapy. The results above conclude that EpCAM can be considered as a potential marker and target for the diagnosis and therapy of GC in the future.