Over-expression of Nectin-4 promotes progression of esophageal cancer and correlates with poor prognosis of the patients

The human EC tissue array (catalog: Heso-Squ172Sur-01) was purchased from Shanghai Outdo Biotech Co., Ltd. (Shanghai, P. R. China). A total of 94 patients who underwent surgery between July 2006 and October 2008 were enrolled in this study. Among these patients, 78 cases of adjacent normal tissues were also included in the tissue array. All these cancer tissues and adjacent normal tissues were confirmed using hematoxylin and eosin (H&E) staining after surgical resection. Incomplete tissue samples and several missing tissue samples were excluded during the heat-induced antigen retrieval, and finally a total of 82 cases were included in the present statistical analysis. Table 1 shows the detailed clinical parameters of the participants. All patients gave informed consent for participation, and the protocol for the present study was approved by the ethics committee of the Third Affiliated Hospital of Soochow University.

The immunostaining of Nectin-4 in EC tissue array was performed according to our published protocol [14, 15]. Briefly, the antigen retrieval was carried out by heating the tissue sections at 100 °C for 30 min in EDTA solution (pH 9.0). Goat anti-human Nectin-4 antibody (AF2659, R&D Systems) was used at the concentration of 10 μg/mL. The Polink-2 plus^® polymer HRP detection system for goat primary antibody (Zhongshan Golden Bridge Biotechnology, Beijing, China) was used according to the manufacturer’s instructions. Diaminobenzene was used as the chromogen, and hematoxylin was employed as the nuclear counterstain. The evaluation of Nectin-4 staining was performed according to the H-score method as described in our published reports [14‐17].

Human EC cell lines Eca-109 and TE-1 were obtained from Chinese Academy of Sciences, Shanghai Institutes for Biological Sciences. The cells were maintained in RPMI-1640 or DMEM supplemented with 10% FBS in the presence of benzylpenicillin (100 U/mL), streptomycin (100 μg/mL) and 2 mM l-glutamine, and were cultured under standard culture conditions (5% CO₂, 37 °C).

The full-length of Nectin-4 (NM_030916.2; GenBank) fragment was synthesized by Sangon Biotech Shanghai Co., Ltd. (Shanghai, China) and cloned to pLV-IRES-ZsGreen-T2A-puro vector (Promega Biotech Co., Ltd., Madison, WI, USA). The small hairpin RNA (shRNA) against the human Nectin-4 gene was obtained from Shanghai GeneChem Co. Ltd (Shanghai, China), and cloned into a lentiviral gene transfer vector pLV-U6-GFP. The shRNA target sequence against Nectin-4 was 5′-CACTCCAAATACGGGCTTCAT-3′. The EC cell lines Eca-109 and TE-1 were transfected with LV-Nectin-4-shRNA, LV-NC, LV-Nectin-4-OE, and LV-Vector-Ctrl, and then selected using puromycin (2 µg/mL; Sigma-Aldrich; Merck Millipore) for more than 2 weeks.

Total RNA was extracted from EC cells, and the RNA quality was determined according to the methods as described in our previous studies [18]. The PCR reactions were performed on an ABI 7600 system (Applied Biosystems, USA) according to the manufacturer’s instructions. Human GAPDH was selected as a housekeeping gene. Primers were synthesized as follows, GAPDH forward primer: 5′-TGACTTCAACAGCGACACCCA-3′, GAPDH reverse primer: 5′-CACCCTGTTGCTGTAGCCAAA-3′; Nectin-4 forward primer: 5′-CTGAGCAGGTTCCCAGGTTT-3′, Nectin-4 reverse primer: 5′-AGAGTTCTTGCCTCTCGCAC-3′. The relative expression of Nectin-4 was calculated by the 2^−ΔΔCT method.

The expression of Nectin-4 at the protein level in different cellular models was determined by Western blotting analysis according to the protocol described in our published reports [14, 15].

The effects of Nectin-4 intervention on biological functions of EC cell lines were assessed according to our previously published protocols [14, 15]. Cell viability was assessed using Cell Counting Kit-8 (CCK-8, Beyotime, Shanghai, China) according to the manufacturer’s instructions. Briefly, 5 × 10³ Eca-109 or TE-1 cells from LV-Nectin-4-shRNA, LV-NC, LV-Nectin-4-OE and LV-Vector-Ctrl were seeded into 96-well plates and incubated for 24, 48 and 72 h. CCK-8 reagent was added to each well at 3 h before the endpoint of incubation, and the absorbance of each well was determined at a wavelength of 450 nm by a microplate reader. An increase or decrease in the absorbance of experimental wells relative to the initial values indicates cell growth or death, respectively. Each experiment was repeated for at least three times.

Eca-109 or TE-1 cells from LV-Nectin-4-shRNA, LV-NC, LV-Nectin-4-OE and LV-Vector-Ctrl groups were cultured in 6-well plates. A small wound area was created using a 200-µL pipette tip when cells reached a 90% confluence. Cells were washed twice with PBS and then incubated in serum-free RPMI-1640 or DMEM medium at 37 °C for 48 h in a 5% CO₂ incubator. Photographs were acquired at two different time points (0 and 24 h). Wound width was measured using a BX50 microscope (Olympus^®) with a calibrated eyepiece grid. Data from three independent experiments were averaged and expressed as a percentage of the original width.

The invasion assay was used to evaluate the effect of intervention of Nectin-4 expression on the invasion ability of human esophageal cancer cells as previously described [12, 13]. Briefly, cells from the different groups were placed in the upper chamber of Matrigel-coated invasion chamber (Corning, NY, USA) and serum-starved for 24 h, and then the medium containing 10% FBS was placed in the lower chamber as a chemo-attractant. After 48 h of incubation, those cells that migrated into the lower chamber were collected and re-suspended, and non-migrating cells were removed from the top of the Matrigel with a cotton-tipped swab. Migrated cells were fixed and stained with 0.1% crystal violet and later photographed under a microscope (Olympus, Tokyo, Japan). Finally, the migrated cells were counted from five randomly selected fields.

The Eca-109 or TE-1 cells from LV-Nectin-4-shRNA, LV-NC, LV-Nectin-4-OE and LV-Vector-Ctrl groups were cultured in 6-well plates and cultured for 48 h. The cells were then washed with ice-cold PBS and fixed in a 70% (v/v) ice-cold ethanol solution at 4 °C overnight. Subsequently, these cells were analyzed by flow cytometry according to the instructions of cell cycle analysis kit (Sigma, MO, USA). The cell cycle information was analyzed using ModFit LT 4.0 software.

The subcutaneous transplantation model study was performed according to our previously published protocol [19]. Briefly, eight groups of female Balb/c nude mice (4–6 weeks old, five mice for each group) were bred under a specific-pathogen-free (SPF) condition with constant humidity and temperature (25–28 °C). Animal protocols were approved by the Animal Care and Use Committee and carried out in compliance with the Guidelines on Animal Welfare of the China National Committee for Animal Experiments. Eca-109 cells (1 × 10⁷) or TE-1 cells (1 × 10⁷) from LV-Nectin-4-shRNA, LV-NC, LV-Nectin-4-OE and LV-Vector-Ctrl groups were suspended in 0.15 mL PBS, and cell suspension was subcutaneously injected into the right back region of nude mice. The tumor size was measured every 2 days with caliper, and the volume was calculated using the formula as follows: Length × Width² × π/6. Growth curves were constructed, and the data were presented as mean ± SEM. Finally, the tumors were harvested from mice at 28 days post injection. The protocol for the tumor model by using nude mice in the present study was approved by the ethics committee of the Third Affiliated Hospital of Soochow University.

Statistical analysis was completed using the paired Student’s t-test, the Wilcoxon signed rank test, the two way ANOVA analysis, the Chi square test or the Log-rank survival analysis where appropriate for final analysis of the data. All the statistical analyses were performed using the GraphPad Prism 5.0 software package (GraphPad Software, Inc., San Diego, USA). A P-value of < 0.05 was considered as statistically significant.

According to the data from https://​www.​proteinatlas.​org, we characterized Nectin-4 expression in different normal human tissues, and the results revealed that Nectin-4 expression could found in normal esophagus tissues at both mRNA (Fig. 1a) and protein (Fig. 1b) levels. Moreover, based on the TCGA data from http://​gepia.​cancer-pku.​cn/​, we studied the mRNA expression levels of Nectin family members including Nectin-1, -2, -3 and -4 in human EC tissues, Fig. 2c shows that the mRNA expression level of Nectin-3 was significantly higher in cancer tissues compared with that in adjacent normal tissues (P < 0.05). Figure 2d illustrates that the mRNA expression level of Nectin-4 was significantly higher in adjacent normal tissues compared with cancer tissues (P < 0.05), while there were no any significant differences in terms of the expressions of Nectin-1 and Nectin-2 between cancer tissues and adjacent normal tissues (Fig. 1a, b). Then we further examined the correlations between the expressions of Nectin-4 in EC tissues and other members namely Nectin-1, -2 and -3. Our results demonstrated that the expression of Nectin-4 at the mRNA level in EC tissues was positively and significantly correlated with that of Nectin-1 (P < 0.0001, Fig. 2e), while it was negatively and significantly correlated with that of Nectin-2 (P < 0.0001, Fig. 2f) and Nectin-3 (P < 0.0001, Fig. 2g). Furthermore, we also examined the prognostic value of Nectin family members in human EC tissues based on TCGA data. We didn’t find any significant correlations between patients’ postoperative survival and Nectin-1 or Nectin-2 (Fig. 2h, i). Figure 2j reveals that the overall survival rate of the patients with higher expression of Nectin-3 at the mRNA level was significantly greater compared with those showing lower Nectin-3 expression (P = 0.006). Moreover, the patients with lower expression of Nectin-4 at the mRNA level exhibited better overall survival rate compared with those showing higher Nectin-4 expression, although the difference was not significant (Fig. 2k, P = 0.266).

In order to further investigate the clinical significance of Nectin-4 expression in human EC, we carried out the immunohistochemistry to study the expression of Nectin-4 in EC tissues. Figure 3a shows that positive staining of Nectin-4 could be found in the cytoplasm and on the membrane of the cancer cells, while weak or negative staining of Nectin-4 could be found in normal esophageal tissues (Fig. 3b). Figure 3c reveals that the staining intensity of Nectin-4 in EC tissues was significantly higher than that in adjacent normal tissues (P < 0.0001). Moreover, the survival analysis showed that the overall survival rate of the patients with higher Nectin-4 expression was significantly poorer compared with those showing lower Nectin-4 expression (HR = 1.704, 95% CI 1.027–2.825, P = 0.039) (Fig. 3d).

Besides, we also analyzed the associations between the expression of Nectin-4 in EC tissues and patient’s clinical parameters. Table 1 shows that the staining intensity of Nectin-4 was positively and significantly associated with tumor size (P = 0.012) and tumor stage (P = 0.016). According to the COX model analysis, the Nectin-4 expression level could serve as an independent prognostic predictor for EC patients (Table 2, HR = 1.795, 95% CI 1.042–3.092, P = 0.035), suggesting that abnormal expression of Nectin-4 was involved in the progression of EC.

In order to further investigate whether the intervention of Nectin-4 expression in human EC cells had effects on cellular functions, we also carried out the cellular study on the knock-down expression or over-expression of Nectin-4 in EC cells. Figure 4 shows that Nectin-4-knockdown expression or Nectin-4-over-expressing cell lines were successfully established using lentiviral transfection, and both real-time RT-PCR and Western blotting analyses were used to confirm the intervention of Nectin-4 expression in human EC cells at the mRNA and protein levels, respectively.

We then examined the effects of the intervention of Nectin-4 expression on the cell proliferation in vitro by using CCK-8 assay in human EC cell lines. As shown in Fig. 5a, in Eca-109 cells, at 48 as well as 72 h after seeding, the proliferation rate of LV-Nectin-4-shRNA group cells was significantly lower than that of LV-NC group cells (both P < 0.01). Figure 5b shows that, in Eca-109 cells, at 48 as well as 72 h after seeding, the proliferation rate of LV-Nectin-4-OE group cells was significantly higher than that of LV-Vector-Ctrl group cells (both P < 0.01). As shown in Fig. 5c, in TE-1 cells, at 24, 48 as well as 72 h after seeding, the proliferation rate of LV-Nectin-4-shRNA group cells was significantly lower than that of LV-NC group cells (P < 0.01 respectively). Figure 5d shows that in TE-1 cells, at 48 as well as 72 h after seeding, the proliferation rate of LV-Nectin-4-OE group cells was significantly higher than that of LV-Vector-Ctrl group cells (both P < 0.01). Moreover, the wound-healing assay was performed to evaluate the effect of the intervention of Nectin-4 expression on the migration ability of human EC cell lines. Knockdown of Nectin-4 expression significantly decreased the cell migration ability of Eca-109 and TE-1 cells, showing that the cell-free area of the LV-Nectin-4-shRNA group was significantly wider than that of the LV-NC group at 24 h (Fig. 6a, in Eca-109, P < 0.05, and in TE-1, P < 0.01). Over-expression of Nectin-4 expression significantly increased the cell migration ability of Eca-109 and TE-1 cells, showing that the cell-free area of LV-Nectin-4-OE group was significantly narrower than that of the LV-Vector-Ctrl group at 24 h (Fig. 6b, both P < 0.01). Furthermore, the transwell invasion assay was performed to evaluate the effect of the intervention of Nectin-4 expression on the invasion ability of human EC cell lines. Knockdown of Nectin-4 expression significantly decreased the number of invaded cells stained with Cristal Violet in the LV-Nectin-4-shRNA group cells compared with the LV-NC group cells (Fig. 7a, in Eca-109, P < 0.01, and in TE-1, P < 0.001). Nectin-4 over-expression significantly increased the number of invaded cells stained with Cristal Violet in the LV-Nectin-4-OE group cells compared with the LV-Vector-Ctrl group cells (Fig. 7b, both P < 0.001).

In order to study the effect of abnormal Nectin-4 expression on the regulation of tumor growth in vivo, we then established the subcutaneous transplantation mouse model. Eca-109 or TE-1 cells of LV-Nectin-4-shRNA, LV-NC, LV-Nectin-4-OE and LV-Vector-Ctrl groups were subcutaneously injected into nude mice. In the tumor model established using Eca-109 cells, the tumor growth curves showed that knockdown expression of Nectin-4 significantly inhibited the tumor growth (Fig. 8a, P < 0.0001), while the tumor growth of Nectin-4 over-expression group also trended to be faster than that of control group (Fig. 8a, P = 0.0831). After 28 days, the tumor weight of the LV-Nectin-4-shRNA group was lighter compared with the LV-NC group (Fig. 8c, P < 0.05), and the tumor weight of the LV-Nectin-4-OE group trended to be heavier compared with the LV-Vector-Ctrl group (Fig. 8c, P < 0.001). In the tumor model established using TE-1 cells, the tumor growth curves showed that knockdown expression of Nectin-4 significantly inhibited the tumor growth (Fig. 8d, P < 0.001), while the tumor growth of Nectin-4 over-expression group also trended to be faster than that of control group (Fig. 8d, P = 0.1816). After 28 days, the tumor weight of the LV-Nectin-4-shRNA group was lighter compared with the LV-NC group (Fig. 8f, P < 0.0001), and the tumor weight of the LV-Nectin-4-OE group trended to be heavier compared with the LV-Vector-Ctrl group (Fig. 8f, P = 0.1314).