Tegaserod maleate was purchased from Selleck Chemicals LLC (Houston, TX, USA). After dissolving in dimethyl sulfoxide to a final concentration of 50 mM, tegaserod maleate was aliquoted and stored at −80°C. Antibodies to detect PEX11B and Ki67 were purchased from Abcam (Cambridge, UK). PEX6 for western blotting was from Signalway Antibody (Pearland, TX, USA). Antibodies to detect PEX13 and PMP70 were from Santa Cruz Biotechnology (Santa Cruz, CA, USA). GAPDH was from Goodhere Biological Technology Co. Ltd. (Hangzhou, Zhejiang, China).

Human ESCC cell lines KYSE150 and KYSE450 were obtained from the Chinese Academy of Sciences (Beijing, China). The Shantou human esophageal epithelium (SHEE) cell line was gifted by Shantou University (Guangdong, China). All cell lines were maintained in RPMI-1640 medium or Dulbecco’s modified Eagle medium supplemented with 10% fetal bovine serum, and cultured at 37°C in a humidified incubator with 5% CO₂.

ESCC cells (KYSE150 cells: 8×10³ cells/well; KYSE450 cells: 1.2×10³ cells/well) were seeded in 96-well plates and cultured with corresponding medium overnight. After incubation for 16 h, cells were treated with tegaserod maleate (0, 3.125, 6.25, 12.5, 25, and 50 μM) for 24 and 48 h. Cells were fixed for 30 min with 4% paraformaldehyde in phosphate buffered saline (PBS) and then stained with 4′,6-diamidino-2-phenylindole (DAPI) at 37°C for 20 min. Cells were counted by using IN Cell Analyzer 6000 software (General Electric Company, GE, USA). The half-maximal (median) inhibitory concentration (IC₅₀) was evaluated by SPSS statistical software, version 21 (IBM Corp., Armonk, NY, USA). Experiments were performed independently three times.

To estimate viability, KYSE150 (2×10³ cells/per well) and KYSE450 (4×10³ cells/per well) cells were seeded into 96-well plates and incubated for 16 h. Cells were exposed to different concentrations of tegaserod maleate (0, 0.25, 0.5, 1, and 2.5 μM) for 24, 48, 72, and 96 h. Cells were fixed by 4% paraformaldehyde for 30 min at 25°C. Then, DAPI was used to stain cells at 37°C for 20 min, and cells were counted using IN Cell Analyzer 6000 software. Experiments were performed independently three times in triplicate.

To assess anchorage-independent cell growth, cells (8×10³ cells/well) suspended in medium were added to 1.25% agar with different concentrations of tegaserod maleate (0, 0.25, 0.5, 1, and 2.5 μM). The cultures were maintained at 37°C in a 5% CO₂ incubator for 2 weeks and then colonies were counted using the IN Cell Analyzer 6000 software. Experiments were performed independently three times.

KYSE150 (200 cells/well) and KYSE450 (500 cells/well) cells were seeded in 6-well plates. Cells were treated with tegaserod maleate (0, 0.25, 0.5, 1, and 2.5 μM) for a designated time and fixed 14 days later with 4% paraformaldehyde for 30 min, and stained with 0.1% crystal violet for 3 min. Colonies containing more than 50 cells were photographed by phone camera, then the number of colonies was measured.

After cells were treated with different concentrations of tegaserod maleate (0, 0.25, 0.5, 1, and 2.5 μM) for 24 h, they were harvested and lysed by radioimmunoprecipitation assay buffer to obtain protein samples. Protein concentration was determined using a bicinchoninic acid protein assay kit (BCA Protein Assay Kit, Beyotime Biotechnology, Shanghai, China). The protein samples were resolved by sodium dodecyl sulfate-PAGE and then transferred onto polyvinylidene fluoride membranes. The membranes were blocked with 5% bovine serum albumin (BSA) or 5% skim milk. Next, the membranes were incubated with primary antibodies overnight at 4°C, followed by incubation with specific secondary antibodies for 2 h at room temperature. Protein bands were visualized using the enhanced chemiluminescence (ECL) detection reagent (Dalian Meilun Biotechnology Co., Ltd, Dalian, China).

KYSE150 and KYSE450 cells were added to 12-well plates at 1.5×10⁵ cells/well and incubated overnight. Cells were treated with tegaserod maleate (0, 0.25, 0.5, 1, and 2.5 μM) for 24 h, fixed with 4% paraformaldehyde, and placed at −20°C overnight. The next day, the cells were permeabilized in 0.5% Triton X‐100/PBS and blocked with blocking buffer (1% BSA-PBS-Tween) for 1.5 h at room temperature. The coverslips or slides were incubated with primary antibodies at 4°C overnight and then with fluorescent secondary antibody for 1.5 h at room temperature. Nuclei were stained using DAPI. Cells or slides were mounted with antifluorescence quencher. Fluorescence images were obtained with IN Cell Analyzer 6000 software and analyzed quantitatively.

The CAT activity in the peroxisomes was detected using a catalase assay kit (Beyotime Biotechnology) according to manufacturer’s instructions. Briefly, under catalysis of peroxidase, residual hydrogen peroxide produces a red product, N-4-antipyryl-3-chloro-5-sulfonate-p-benzoquinonemonoimine, that has an absorption maximum at 520 nm. CAT activity was calculated from the assay results.

Production of ROS was monitored by using a Reactive Oxygen Species Assay Kit (Beyotime Biotechnology) according to the manufacturer’s protocol. KYSE150 and KYSE450 cells treated with tegaserod maleate (0, 0.25, 0.5, 1, and 2.5 μM) were incubated with 10 mM 2,7-dichlorofluorescein diacetate (DCFH-DA) at 37°C in a humidified incubator with 5% CO₂ for 20 min. Fluorescence images and DCF fluorescence intensities were analyzed using the IN Cell Analyzer 6000.

The tissue microarray included tumors and adjacent normal tissues from 114 cases of ESCC (Shanghai Xinchao Biotechnology Co. Ltd., Shanghai, China). PEX13 and PEX11B were detected according to standard immunohistochemical methods, and their expression was evaluated and scored on the basis of staining intensity and area. The intensity was scored as follows: 0, negative; 1, weak; 2, moderate; 3, strong. All tissues were photographed using a microscope camera and analyzed by TissueFAXS Viewer software program (TissueGnostics, Austria).

PEX11B and PEX13 were deleted in ESCC cells by using the CRISPR/Cas9 system. Following the manufacturer’s suggested protocols, viral vectors and packaging vectors were transfected into 293T cells by using Jet Primer (ThermoFisher Scientific, Waltham, MA, USA). After transfection, the medium was changed at 4 h, and cells were cultured for 24 and 48 h. Viral particles were harvested by filtration using a 0.22-mm syringe filter, and then combined with 8 μg/mL polybrene and used to infect KYSE150 cells. KYSE150 cells were selected with 2 μg/mL puromycin for 72 h. Knockout efficacy was determined by western blotting. The oligonucleotide sequences of PEX11B and PEX13 single guide (sg) RNA were designed from an online CRISPR/Cas9 tool (CHOPCHOP: https://chopchop.cbu.uib.no/) (12) and were listed as Table 1 .

To examine the effect of tegaserod maleate on ESCC PDX tumor growth, severe combined immunodeficiency (SCID) mice (6 to 8 weeks old; Vital River Labs, Beijing, China) were used. ESCC tissues (named LEG34) were cut into pieces and planted into the back of the neck of the mice. Mice were randomly divided into three groups (7 animals each) as follows: the vehicle group (normal saline group), 2 mg/kg tegaserod maleate group, and 10 mg/kg tegaserod maleate group. The tegaserod maleate or vehicle was orally administered once every day. Mice were monitored until the average tumor volume in the vehicle group reached a volume of 1000 mm³, at which time treatment was stopped and the tumor tissue was harvested. The tumor growth inhibition (TGI) rate was determined according to the calculation formula, as follows: TGI=100[1-(Treatment final volume-Treatment initial volume)/(Control final volume-Control initial volume). All animal assays were performed under the supervision of the Zhengzhou University Institutional Animal Care and Use Committee (Zhengzhou, Henan, China). All protocols were approved by the Research Ethics Committee of Zhengzhou University.

Tumor tissues from mice were embedded in paraffin blocks for immunohistochemical staining and analysis. Tissues sections were deparaffinized and hydrated. Sections were incubated with 3% H₂O₂ for 10 min and then incubated at 4°C with primary antibodies overnight. Secondary antibody was added and incubated at 37°C for 15 min. Tissue sections were stained using diaminobenzidine and hematoxylin. Finally, sections were dehydrated and covered with glass slides. All tissue sections were photographed using a microscope camera and analyzed using TissueFAXS Viewer software program.

SPSS statistical software, version 21 (IBM Corp.) was used for all statistical analyses. The results were analyzed via Student’s t-test or one-way analysis of variance. Quantitative data were expressed as mean values ± standard deviations. Statistical significance was defined by a p-value < 0.05.

To identify drugs with anti-esophageal cancer activity, we screened FDA-approved drugs and found tegaserod maleate had an inhibitory effect and resulted in a cell viability rate of only 0.29% in KYSE450 cells at 50 μM ( Figure 1A ). The chemical structure of tegaserod maleate is shown in Figure 1B . Through a cytotoxicity assay ( Figure S1 ), we found that the half-maximal inhibitory concentration (IC₅₀) value in KYSE150 cells was 6.867 μM at 48 h and that in KYSE450 cells was 4.476 μM ( Figure 1C ). The IC₅₀ in the immortalized human esophageal epithelial cell line SHEE was 9.688 μM ( Figure S2 ). These results indicated that tegaserod maleate had a significant inhibitory effect on KYSE150 and KYSE450 cells (p < 0.05) but a less effect on SHEE cells ( Figure 1D ). We then performed a proliferation assay ( Figure 1E ), an anchor-independent cell growth assay ( Figure 1F ), and a colony formation assay ( Figure 1G ). We found that tegaserod maleate significantly inhibited anchorage-dependent growth and anchorage-independent growth of ESCC cells in a dose-dependent manner. Together, these results confirmed that tegaserod maleate could inhibit the proliferation of ESCC cells in vitro.

To explore the mechanism by which tegaserod maleate inhibits ESCC proliferation in vitro, we performed a proteomics analysis. Cells were harvested for mass spectrometry analysis after a 24-h treatment with tegaserod maleate. The process is illustrated in Figure 2A . In total, compared with an untreated control group, 6401 differential proteins were detected, of which 5131 had quantitative information. After normalization, 161 differentially expressed proteins were detected ( Figure 2B ). Of these proteins, 97 were upregulated and 64 were downregulated (fold change >1.5, p < 0.05) ( Figure 2C ). The quantitative values of all differentially expressed proteins were normalized using heat maps for downregulated proteins ( Figure 2D ) and upregulated proteins ( Figure S3 ). We ranked the downregulated differentially expressed proteins and displayed the top 10 in accordance with the enrichment score ( Figure 2E ). The downregulated proteins were enriched in the Kyoto Encyclopedia of Genes and Genomes (KEGG) database. The top five pathways were displayed from small to large according to Fisher’s exact test p-value, and only the first two pathways were statistically significant (p < 0.05). The KEGG database showed that drug-altered proteins, including PEX6, PEX11B, PXMP4, and PEX13, were closely related to the peroxisome signaling pathway ( Figure 2F ). The spectra of these proteins were represented by the highest specific peptide score ( Figure S4 ). With a confidence level >0.7 as the threshold, only three differentially expressed proteins, namely, PEX6, PEX11B, and PEX13 interacted in the peroxisome pathway, in accordance with data in the STRING database ( Figure 2G ). Western blotting showed that levels of PEX6, PEX11B, and PEX13 were decreased in KYSE150 cells treated with tegaserod maleate ( Figure 2H ). Moreover, using peroxisome integral membrane protein PMP70 as the marker protein of peroxisome, we found that PMP70 was downregulated in tegaserod maleate-treated KYSE150 cells through western blotting ( Figure S5A ). These findings indicated that peroxisomes and levels of peroxisome-related proteins were downregulated following tegaserod maleate treatment.

Peroxisomes play an important role in growth and cell balance (13). Therefore, we probed whether tegaserod maleate would affect peroxisome function. An immunofluorescence assay indicated that the fluorescence intensity of the peroxisome integral membrane protein PMP70 decreased by 18.4% in KYSE150 cells ( Figure 3A ) and by 17.2% in KYSE450 cells ( Figure 3B ) treated with 2.5 μM of tegaserod maleate. Similarly, we assayed the activity of catalase, a marker enzyme of peroxisomes, and showed that tegaserod maleate attenuated catalase activity in KYSE150 and KYSE450 cells. This inhibition showed concentration dependence ( Figure 3C ). Consistent with previous results, the level of ROS in ESCC cells increased after tegaserod maleate treatment, presumably because the peroxisomes with attenuated catalase activity were unable to eliminate intracellular hydrogen peroxide ( Figure 3D ). Additionally, we verified, by western blotting, that protein levels of PEX6, PEX11B, and PEX13 decreased with increasing tegaserod maleate concentration ( Figure 3E ). Collectively, these results indicated that tegaserod maleate downregulates peroxisome function.

PEX11B and PEX13 are important membrane components of peroxisomes: PEX11B participates in the elongation of peroxisomes (14) and PEX13 is involved in the transport of matrix enzymes (15). We searched for TCGA gene expression data in UALCAN (16) and found that PEX11B and PEX13 were upregulated in ESCC ( Figures 4A, C ). Subsequently, we used Kaplan Meier Plotter (17) to observe that high expression of PEX11B and PEX13 in ESCC patients predict poor overall survival (OS) and recurrence-free survival (RFS) ( Figures 4B, D ). On comparing expression of PEX11B and PEX13 between ESCC samples and normal tissues using a tissue microarray, PEX11B and PEX13 were found to be highly expressed in ESCC ( Figures 4E, G ) and correlated with clinicopathological grade ( Figures 4F, H ). In addition, we also examined the expression of PEX6, but TCGA database showed that PEX6 was not highly expressed in ESCC and was not associated with a poor prognosis ( Figure S6 ).

We further explored the role of PEX11B and PEX13 in tumor suppression. First, we established KYSE150-stably transfected cell lines with a plasmid expressing sgRNA against PEX11B (sgPEX11B) and or a plasmid expressing sgRNA against PEX13 (sgPEX13) respectively and a control sequence (sgControl). We verified the knockout effect of PEX11B and PEX13 by western blotting, respectively ( Figure 5A ). Subsequently, we performed a cell proliferation assay ( Figure 5B ) and a colony formation assay ( Figure 5C ). The results showed that cell proliferation was significantly inhibited after silencing PEX11B or PEX13. On the basis of these results, we hypothesize whether the inhibitory effect of tegaserod maleate depends on the expression of PEX11B or PEX13. Therefore, we treated PEX11B and PEX13 knockout KYSE150 cells respectively with tegaserod maleate. After 96 h, cell viability indicated that knocking out PEX11B or PEX13 in KYSE150 cells were less sensitive to different concentrations of tegaserod maleate than sgControl group ( Figure 5D ). In summary, we concluded that the inhibitory effect of tegaserod maleate depended on PEX11B and PEX13 protein levels.

The PDX model was used to further evaluate the antitumor activity of tegaserod maleate in vivo. A ESCC patient’s tumor (female, moderately differentiated squamous cell carcinoma of origin, which was named LEG34) tissue was implanted into the back of SCID (severe combined immunodeficiency) mice, thereafter, tegaserod maleate (2 or 10 mg/kg) was administered to the mice daily by gavage ( Figure 6A ). Compared with the vehicle group, treatment with 2 or 10 mg/kg tegaserod maleate had no effect on the body weight of the mice ( Figure 6B ) but significantly inhibited tumor growth ( Figures 6C, D ). We calculated tumor weight ( Figure 6E ) and found that tumor growth inhibition (TGI) reached 55.17% with 2 mg/kg and 66.11% with 10 mg/kg tegaserod maleate ( Figure 6F ). Subsequently, we observed the growth of individual tumors in the different groups (vehicle, 2, or 10 mg/kg tegaserod maleate), and found that tegaserod maleate inhibited tumor growth ( Figures 6G-I ). We used immunohistochemistry to analyze the expression of Ki67 in tumor tissues ( Figure 6J ). Simultaneously, we detected the expression of PEX11B, PEX13 ( Figure 6K ) and PMP70 ( Figure S5B ) in tumor tissues by western blotting. Peroxisome and expression of all these proteins was suppressed in the tegaserod maleate-treatment group compared with that in the vehicle group.

Through the above studies, we believe that tegaserod maleate restricts the function of peroxisomes by downregulating PEX11B and PEX13 in the peroxisome signaling pathway. Weakened catalase is unable to remove hydrogen peroxide in the peroxisome, resulting in the increase of ROS in cells ( Figure 7 ). Therefore, tegaserod maleate could inhibit ESCC proliferation both in vitro and in vivo.