Vascular endothelial zinc finger 1 (VEZF1; cat. no SC-365560; Santa Cruz Biotechnology, Inc.) β-actin (cat. no GT5512; Genetex International Corporation), and versican (VCAN) antibody (cat. no SAB1408906; MilliporeSigma) were used. Recombinant human visfatin (cat. no. 130-09-25UG) was obtained from PeproTech, Inc. The small interfering (si)RNAs targeting VEZF1 (cat no. sc-94046) and VCAN (cat no. sc-41903) were purchased from Santa Cruz Biotechnology, Inc. A non-targeting negative control siRNA (cat no. D-001810-10-05) was purchased from Thermo Fisher Scientific, Inc. PI3K (Ly294002) inhibitor (cat. no. ALX-270-038) was obtained from Enzo Life Sciences, Inc. AKT (cat. no. A6730) and mTOR (rapamycin) inhibitors (cat. no. R0395) were obtained from Sigma-Aldrich (Merck KGaA). The miR-3613-5p mimic (5′-UGUUGUACUUUUUUUUUUGUUC-3′) and miR mimic negative control (5′-UUGUACUACACAAAAGUACUG-3′) were purchased from AllBio. All other reagents and chemicals were obtained from MilliporeSigma, unless otherwise specified.

The invasive ESCC cell line KYSE-410 (cat no. 94072023) was obtained from the European Collection of Cell Cultures and human ESCC cell line CE81T (cat. no. 60166) was purchased from Bioresource Collection and Research Centre. Cells were cultured in either RPMI-1640 or DMEM (Gibco; Thermo Fisher Scientific, Inc.) supplemented with 10% heat-inactivated fetal calf serum (Corning, Inc), 2 mM L-glutamine, 100 U/ml penicillin and 100 µg/ml streptomycin. All cells were maintained at 37°C in a humidified atmosphere containing 5% CO₂.

For miRNA sequencing (NEXTflex small RNA sequencing kit v3, cat no. NOVA-5132-06, PerkinElmer) was utilized, high-quality total RNA samples from the visfatin (30 ng/ml)-treated for 24 h at 37°C and untreated control KYSE-410 cells were utilized. The experimental workflow was performed by Azenta Life Sciences. To prepare the library, 1 µg total RNA was used quantified using both Qubit and qPCR methods. The qualified library were sequenced pair end PE150 (150 pair-End) Small RNA sequencing was conducted using the Illumina, Inc. HiSeq/Novaseq or MGI2000 platform. Total RNA was extracted using the TRIzol (Thermo Fisher Scientific, Inc.) and RNA integrity was confirmed with an Agilent 2100 Bioanalyzer (RNA integrity number >7). To prevent adaptor-dimer formation, an excess of 3′ SR Adaptor for Illumina was hybridized with the SR RT Primer. Subsequently, the 5′ SR Adaptor for Illumina was ligated to the small RNA using a 5′ Ligation Enzyme. First-strand cDNA synthesis was performed using ProtoScript II Reverse Transcriptase (New England Biolabs, US). Each sample was then amplified by PCR kit (Bioo Scientific, PerkinElmer) using thermocycling conditions: initial denaturation 95°C for 2 min, 20–25 cycles of denaturation 95°C for 20 sec, annealing 60°C for 30 sec, Extension 72°C for 15 sec, final extension 95°C for 2 minusing P5 (Forward, 5′-GTTCAGAGTTCTACAGTCCGACGATC-3′) and P7 (Reverse, 5′-AGATCGGAAGAGCACACGTCT-3′) primers and the PCR product was purified by DNA Clean Beads (Bioo Scientific, PerkinElmer). The purified products of 140–160 bp were recovered and cleaned using PAGE (6%) and validated using an Agilent 2100 Bioanalyzer. The raw readings underwent quality control before processing, which included removing adapter sequences and contaminants. To guarantee data quality, analysis of the lengths and counts of the filtered reads was performed by Trimmomatic (V0.30), Cutadapt (V1.3) and FastQC (V0.10.1), along with an evaluation of the data volume (9). DEseq2 (V1.6.3), DEseq (V1.18.0), EdgeR (V3.4.6) and bowtie2 (V2.1.0), software was used for Heatmap and volcano plot to analyzed differentially expressed miRNA.

The levels of visfatin in patients with primary and metastatic esophageal cancer were analyzed using a UALCAN, dataset and cell motility genes associated with VEZF1 were obtained from The Cancer Genome Atlas (TCGA) database (3,11). Target genes were predicted using TargetScan (targetscan.org/vert_80/), miRTarBase (https://mirtarbase.cuhk.edu.cn), miRDB (https://mirdb.org/) and ENCORI (https://rnasysu.com/encori/index.php) databases (24). Spearman correlation was used to analyze gene correlation. Gene expression levels in ESCC patients were analyzed using the GEO dataset GSE161533. Kaplan-Meier analysis was performed to assess the VCAN levels in ESCC. Expression levels of Visfatin, VEZF1, and VCAN were evaluated using GSE77861, while miR-3613-5p levels were examined using GSE97051 in ESCC patient tissues.

Transwell inserts (Costar, Inc.; 8-µm pores) were used in 24-well plates for the migration experiments and pre-coated with a layer of Matrigel at 37°C for 30 min before the invasion assay. KYSE410 and CE81T Cells were pretreated with or without (10 µM) of PI3K (Ly294002), AKT inhibitor (AKTi), mTOR (Rapamycin) inhibitors. KYSE410 and CE81T cells were transfected using (Lipofectamine 2000, Invitrogen, Thermo Fisher Scientific, Inc.) with or without siVEZF1, siVCAN siRNAs and miRNA mimics (50 nM) and incubated at 37°C for 24 h, immediately followed by migration and invasion assay. The upper chamber contained 1×10⁴ KYSE410 and CE81T cells in 200 µl serum-free medium (DMEM or RPMI), whereas the lower chamber contained 300 µl 10% FBS and medium with various concentrations (1,3,10, 30) ng/ml of visfatin and incubated for 24 h at 37°C in 5% CO₂. Cotton-tipped swabs were used to remove the Matrigel from the upper side of the filters, and PBS was used to wash the filters (23,24). Cells were fixed in 3.7% formaldehyde for 5 min at room temperature, incubated for 24 h at 37°C in 5% CO₂ and stained with 0.05% crystal violet in PBS for 20 min at room temperature. Stained cells were observed under an Olympus CKX53 inverted light microscope, and quantified by ImageJ (V1.52a; imagej.net/ij/) and GraphPad prism (V8.0; graphpad.com/guides/prism/8/user-guide/index.htm).

TRIzol (Thermo Fisher Scientific) cat no. 12183555) was used to extract total RNA from the esophageal cancer KYSE410 cells. In summary, oligo-dT primers (Thermo Fisher Scientific, Inc.), 10 mM dNTP (Cyrusbioscience), 5× standard buffer (Invitrogen; Thermo Fisher Scientific) and 0.1 M DTT (Invitrogen; Thermo Fisher Scientific) were used to reverse-transcribe 1 µg RNA into cDNA in compliance with the manufacturer's instructions. The KAPA SYBER FAST qPCR kit (Applied Biosystems; Thermo Fisher Scientific, Inc.) was used to mix 100 ng cDNA sample with specific primers. qPCR was performed using a Senso Quest Labcycler thermal cycler. The thermocycling conditions were as follows: Initial denaturation at 95°C for 6 min, followed by 40 cycles of denaturation at 95°C for 30 sec, annealing at 60°C for 30 sec and extension at 72°C for 1 min, with a final extension at 72°C for 10 min. The Mir-XTM miRNA First Strand Synthesis kit (Takara Bio Inc.) was used to create cDNA from 100 ng total RNA for the miRNA assay. The endogenous control GAPDH was used to achieve relative quantification of gene expression. The comparative Cq approach was performed to calculate the relative expression (25,26). The sequences of primers were as follows: miR-3613-5p, 5′-TGTTGTACTTTTTTTTTTGTTC-3′ (melting temperature (Tm), 43.7°C; length, 22 bases); VEZF1: Forward, 5′-GGTTCTGCAGCATTTCACCC-3′ and reverse, 5′-TGATGGGAAGCTTCATGGGC-3′ (Tm, 53.8°C; length, 20 bases each); VCAN: Forward, 5′-GTAACCCATGCGCTACATAAAGT-3′ (Tm, 53.5°C; length, 23 bases) and reverse, 5′-GGCAAAGTAGGCATCGTTGAAA-3′ (Tm, 53.0°C; length, 22 bases) and GAPDH: Forward 5′-ACCACAGTCCATGCCATCAC-3′ (Tm, 53.5°C, length, 20 bases) and reverse, 5′-TCCACCACCCTGTTGCTGTA-3′ (Tm, 53.8, length, 20 bases). Relative gene expression levels were calculated using the 2^−ΔΔCq method (27).

Proteins from KYSE410 cells were extracted using RIPA lysis buffer (cat. No. P0012, Beyotime Institute of Biotechnology). The total protein concentration was determined using a BCA Protein Assay Kit and 30 µg of protein per lane was used for analysis. Proteins were separated by SDS-PAGE using a 10% gel and transferred onto PVDF membranes (Merck KGaA). Membranes were blocked with 5% non-fat milk at room temperature for 1 h and incubated with primary antibodies VEZF1 (1:1,000), VCAN (1:1,000) as target proteins and β-actin (1:3,000) for an entire night (18–20 h) at 4°C. Membranes were incubated for 1 h at room temperature with horseradish peroxidase-conjugated secondary antibodies (1:5,000) goat anti-mouse IgG (cat. no. sc-516102), Santa Cruz Biotechnology, Inc. An ECL kit (MilliporeSigma) was used to detect the expression of the target protein and an ImageQuant LAS 4000 biomolecular imager was used for visualization (28,29).

A luciferase assay kit was utilized to track the luciferase activity in order to measure the 3′-untranslated region (UTR). After transfecting the cells with either the wild-type (wt)- or mutant (mt)-VEZF1-3′-UTR luciferase plasmid (Stratagene; Agilent Technologies, Inc.), using Lipofectamine 2000 (Invitrogen; Thermo Fisher Scientific) the cells were transfected for 24 h at 37°C using miR-3613-5p mimic (sequence: 5′-ACAAAAAAAAAAGUACAACAUU-3′; (AllBio). Following 24 h transfection, cells were lysis and instantly the luciferase activity was measured using the Dual-Luciferase^® Reporter Assay System (Promega Corporation). Firefly luciferase activity was normalized to Renilla luciferase activity to control for transfection efficiency.

Data were analyzed using ImageJ software (V1.52a) (https://imagej.net/ij/) and GraphPad Prism software (V8.0) (graphpad.com/guides/prism/8/user-guide/index.htm). All data are presented as the mean ± SD from 3 independent experiments. Statistical significance between two groups was assessed using the unpaired Student's t-test. Comparisons involving >2 groups with a single variable were performed using one-way ANOVA followed by Bonferroni's post hoc test. P<0.05 was considered to indicate a statistically significant difference.

Visfatin is key for the development of numerous types of cancer (3,30). However, the mechanisms through which visfatin affects esophageal cancer metastases remain unknown. The UALCAN data revealed that the levels of visfatin were higher in patients with metastatic esophageal cancer than in those with primary esophageal cancer (Fig. 1A). Transwell migration assay was used to examine the effects of visfatin on the migration of esophageal cancer cells. Visfatin promoted the migration of KYSE-410 esophageal cancer cells in a concentration-dependent manner (Fig. 1B). Additionally, visfatin enhanced the invasive ability of esophageal cancer cells (Fig. 1C).

The metastasis of esophageal cancer is associated with dysregulated miRNAs (31). To investigate whether miRNAs mediate visfatin-induced esophageal cancer cell migration, miRNA sequencing was performed to assess the differential expression of miRNAs in KYSE-410 cells treated with visfatin. The resulting heatmap and volcano plot illustrated differentially expressed miRNAs (Fig. 2A and B); among these, miR-3613-5p was the most downregulated (Fig. 2B). Visfatin (≥3 ng/ml) inhibited miR-3613-5p synthesis in a concentration-dependent manner (Fig. 2C). Transfection with a miR-3613-5p mimic increased the miR-3613-5p expression compared with control (Fig. 2D). miR-3613-5p mimic antagonized the visfatin-induced promotion of cell migration and invasion (Fig. 2E and F). Consistent with the KYSE-410 results, miR-3613-5p mimic treatment inhibited visfatin-induced migration and invasion in CE81T cells (Fig. 2G and H). Thus, these findings demonstrate that visfatin promoted esophageal cancer cell migration by decreasing miR-3613-5p production.

A total of four publicly available miRNA databases (TargetScan, miRTarBase, miRDB and ENCORI) predicted that miR-3613-5p targets seven potential candidates (Fig. 3A). Among these, VEZF1 was markedly upregulated compared with ANP32B in patients with esophageal cancer (Fig. 3B). Additionally, VEZF1 was more highly upregulated in patients with metastatic than in those with primary esophageal cancer, however this was not significant (Fig. 3C). The direct application of visfatin to esophageal cancer cells enhanced VEZF1 mRNA and protein expression (Fig. 3D and E). Transfection of the cells with VEZF1 siRNA, effectively suppressed VEZF1 protein levels, as confirmed by western blot analysis (Fig. 3F), and inhibited the visfatin-induced promotion of cell migration and invasion (Fig. 3G-I). These findings were validated in CE81T cells, where VEZF1 siRNA also reversed visfatin-induced cell motility, confirming a consistent effect across ESCC cell lines (Fig. 3J-L). Subsequently, wt- and mt-VEZF1-3′-UTR luciferase plasmids were generated to examine the direct binding effects of miR-3613-5p on VEZF1 (Fig. 4A). The miR-3613-5p mimic decreased wt-VEZF1-3′-UTR, but not mt-VEZF1-3′-UTR luciferase activity (Fig. 4B). Furthermore, the miR-3613-5p mimic blocked visfatin-induced VEZF1 mRNA expression (Fig. 4C) and protein levels (Fig. 4D).

To investigate whether the visfatin-induced expression of VEZF1 regulates cell motility genes, TCGA database was searched for genes associated with VEZF1 (Fig. 5A). Among 300 genes exhibiting a positive correlation, 13 genes were also associated with cell adhesion functions (Fig. 5B). Data from the GSE161533 database indicated that VCAN was the most upregulated gene in patients with esophageal cancer (Fig. 5C). Kaplan-Meier analysis confirmed that the VCAN levels were higher in patients with esophageal cancer than in healthy controls (Fig. 5D). Visfatin promoted the mRNA and protein expression of VCAN (Fig. 6A and B). Transfection with VCAN siRNA effectively decreased VCAN protein levels, as confirmed by western blot analysis (Fig. 6C), and also attenuated visfatin-induced cell motility (Fig. 6D-F); this effect was validated in CE81T cells (Fig. 6G-I). miR-3613-5p mimic and VEZF1 siRNA also suppressed visfatin-induced VCAN expression (Fig. 6J and K), indicating that the inhibition of miR-3613-5p and the promotion of VEZF1 occurred upstream of visfatin-induced VCAN expression and cell motility. Mechanically, visfatin inhibits miR-1264 and promotes platelet derived growth factor C (PDGF-C) synthesis through activation of the PI3K/AKT/mTOR signaling pathway (19). To elucidate the molecular mechanism, the present study examined the role of the PI3K/AKT/mTOR pathway in visfatin-mediated ESCC migration and invasion Transwell assay results indicate that treatment with PI3K (Ly294002), AKTi and mTOR (rapamycin) pathway inhibitors effectively reversed the migration and invasion effects induced by visfatin, confirming the involvement of PI3K/AKT/mTOR signaling axis in visfatin-driven migration and invasion (Fig. 6L-Q). To explore the clinical relevance of visfatin-related genes, GSE77861 dataset was analyzed to compare gene expression between normal and esophageal cancer tissue. Visfatin (Fig. 6R), VEZF1 (Fig. 6T) and VCAN (Fig. 6U) were significantly upregulated in esophageal cancer tissues compared with normal controls (P<0.05). In addition, analysis of the GSE97051 dataset showed that miR-3613-5p expression was slightly decreased in cancerous tissue compared with normal samples (Fig. 6S). Together, these findings support the potential involvement of the visfatin/miR-3613-5p/VEZF1/VCAN axis in the progression of esophageal cancer.