The expression and the tumor suppressor role of CLDN6 in colon cancer

As a member of the tight junction family, CLDN6 is a tumor suppressor in breast cancer, but its role in colon cancer is unknown. In this research, we aimed at revealing the function of CLDN6 in colon cancer. We found that colon cancer tissues lowly expressed CLDN6, and the expression of CLDN6 was negatively correlated with lymph node metastasis. Similarly, CLDN6 was lowly expressed in the colon cancer cell line SW1116, and overexpression of CLDN6 inhibited cell proliferation in vitro and in vivo. Consistently, the migration and invasion abilities of cells were significantly inhibited after CLDN6 overexpression. In addition, we demonstrated that CLDN6 may inhibit the migration and invasion abilities by activating the TYK2/STAT3 pathway. Therefore, our data indicated that CLDN6 acted as a tumor suppressor and had the potential to be regarded as a biomarker for the progression of colon cancer.


Introduction
Colon cancer is the fourth most deadly cancer with about 900,000 deaths every year in the world [1]. Approximately 25% of colon cancer patients have hepatic metastasis on initial diagnosis, and about 50% would develop hepatic metastasis within 3 years of the primary surgery [2]. Therefore, searching for new biomarkers related to tumor development and clarifying underlying molecular mechanisms could be critical for screening patients at high risk of recurrence and identifying novel therapeutic targets.
Claudins (CLDNs) are major tight junction proteins and the alteration of CLDNs is one of the mechanisms for the loss of cell adhesion, which is an important step in tumor metastasis [3]. In addition, CLDNs regulate cancer cells proliferation [4], apoptosis [5,6], autophagy [7][8][9], drug resistance [10], and so on. Several CLDNs are associated with the malignant phenotypes of colon cancer. Studies have shown that CLDN1 increased the metastatic behavior of colon cancer by the promotion of EMT [11,12]. On the contrary, Resnick et al. [13] found that lower expression of CLDN1 was associated with higher tumor grade, lymphovascular invasion and poor survival in stage II colon cancer. Loss of CLDN3 increased the abilities of migration and invasion of colon cancer cells [14], and CLDN7 correlated with venous invasion and liver metastasis of colon cancer [15]. However, the role and mechanism of CLDN6 in colon cancer progression were unclear.
The protein tyrosine kinase 2 (TYK2) is the first identified member of the Janus kinase (JAK) family and is aberrantly activated in various types of cancer [16]. TYK2 is required to transduce the signaling from growth factors, cytokines, and oncogenes to the signal transducers and activators of transcription (STAT) family, including STAT3. STAT3 is an important transcription factor that plays a crucial role in the development of cancer by regulating the expression of genes related to the cell cycle, motility, apoptosis, and metastasis [17]. For that, it is regarded as an oncogene. However, studies have found that STAT3 could be either an oncogene or tumor suppressor under different conditions [18].
Signaling transduction involving CLDN6 has received much attention in recent years. There is a PDZ binding motif at the carboxy terminus of CLDN6 that can bind to proteins that contain PDZ domains [19]. In this way, CLDN6 is involved in cell signaling transduction that regulates 1 3 multiple biological behaviors of tumors. The signaling pathways involved in tumors that CLDN6 participates in include but are not limited to the SFK/PI3K/AKT pathway [20], the AF6/ERK pathway [21], and the ASK1/p38/JNK pathway [22] in breast cancer, the YAP/Snail pathway in gastric cancer [23], and the PI3K/AKT/mTOR pathway in endometrial carcinoma [24]. We previously performed RNA sequencing of MCF-7/CLDN6 cells and found that the differentially expressed genes were enriched in the TYK2/STAT3 pathway. Therefore, we hypothesized that CLDN6 may regulate the TYK2/STAT3 pathway in colon cancer.
In this research, we examined the expression of CLDN6 in colon cancer tissues and analyzed the relationship between CLDN6 expression and clinicopathological characteristics of colon cancer patients. Also, we investigated the effect of CLDN6 on the biological characteristics of colon cancer cells. Moreover, we explored the potential mechanism by which CLDN6 regulated the migration and invasion abilities of colon cancer cells. Our study suggested that CLDN6 might be a useful marker to identify the progression of colon cancer.

Clinical samples and data collection
Paraffin sections were collected from 107 patients pathologically with colon cancer who were treated with surgery at the Eastern Division of the First Hospital of Jilin University from January 2012 to June 2013. This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethics Committee of Jilin University and informed consent was obtained from all patients. There were 66 males and 41 females with an average age of 58.3 years old. Among them, 11 tumors were well-differentiated, 65 tumors were moderately-differentiated and the other 31 tumors were poorly-differentiated. There were 49 cases with lymph node metastasis and 58 cases without metastasis. 107 cases of normal colonic tissues adjacent to the cancer were taken as control specimens.

Immunohistochemistry (IHC)
IHC was performed as described [25]. Tissue sections were immunostained with CLDN6 antibody (Santa Cruz, CA, USA). Diaminobenzidine (DAB) was used for color development. CLDN6 expression was indicated in brown and was expressed in the cell membrane. Scoring was performed as follows: negative, < 10% positive tumor cells; positive, ≥ 10% positive tumor cells. Positive or negative reactions were determined in five random fields of each sample.

Plasmid and transfection
A CLDN6 overexpression plasmid was established as previously described [26]. Cells were transfected with plasmid by using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA), following the procedure recommended by the manufacturer. The single clone was screened out by G418 and then cultured for amplification.

Western blot (WB) analysis
RIPA lysis buffer (Beyotime, Shanghai, China) was used to extract protein from cells. The protein concentration was determined by a BCA protein assay kit (Beyotime). 40 μg of each sample were applied to 12% SDS-polyacrylamide gel, then proteins were transferred onto a nitrocellulose membrane (Millipore, California, USA) and incubated with each primary antibody overnight at 4 °C, followed by incubation for 1 h at room temperature with HRP-conjugated secondary antibody.

Colony formation assay
Single-cell suspensions were prepared and plated into sixwell plates. The visible colonies appeared after 3 weeks and were fixed with methanol for 20 min, and then stained with Giemsa solution for 5 min. Cell colonies with a diameter > 0.5 mm were counted and the colony formation 1 3 efficiency was calculated with the following formula: % of colony formation = (number of colonies formed/number of cells inoculated) × 100%.

Cell viability assay
The viability of cells was monitored using the Cell Counting Kit-8 (CCK-8) assay (Dojindo, Kumamoto, Japan) according to the manufacturer's instructions. The optical density (OD) at 450 nm of each group was measured at 12, 24, 48,72, and 96 h on a Microplate Reader (Thermo, Schwerte, Germany).

Subcutaneous xenograft models
A total number of twelve 6-week-old BALB/c nude mice (body weight: 16-20 g) were randomly divided into two groups and subcutaneously injected with SW1116/vec or SW1116/CLDN6 cells (5 × 10 6 cells per mouse) into the hind flank. Calipers were used to measure the length (L) and width (W) of the tumors every 4 days. The tumor volume was estimated by the formula V = 0.52 × L × W 2 . All animal experiments and their care were approved by the Experimental Animal Ethical Committee of Jilin University and were carried out following relevant institution guidelines.

Wound-healing assay
When the cell grew to about 80% confluence, a single scratch was produced by dragging a 10 μL pipette tip. Then these cells were washed with PBS three times to remove cell debris. The width of the wounded areas was measured by microscopy (OLYMPUS, Tokyo, Japan) at 0 h (W0) and 24 h (W24) after injury. The migration rate was calculated as (W0 − W24)/W0.

Transwell chamber assay
Matrigel invasion assay was performed using Transwells containing 8.0 μm pore membranes (Corning, Lowell, MA, USA). Cells were placed in the upper chamber of the Transwell for 48 h. Then the chambers were washed twice with PBS. The filter side of the upper chamber was cleaned with a cotton swab. Next, the remained cells in the upper chamber of the membrane were cut out of the insert. Cells were fixed in methanol and stained with 5% Giemsa for 30 min at room temperature. The invaded cells were photographed and counted under a light microscope.

Statistical analysis
Statistical analysis was performed using GraphPad Prism 8.0 (GraphPad, San Diego, CA, USA). All data were presented as mean ± SD. Statistical significance was determined by the Student's t-test or one-way analysis of variance (ANOVA). The protein expression levels and clinicopathologic features were compared by the chi-square test. Differences between groups were considered statistically significant if P < 0.05.

CLDN6 is lowly expressed in colon cancer
To detect the expression of CLDN6 in colon cancer, we performed an IHC assay using colon cancer tissues. As shown in Fig. 1a, positive expression of CLDN6 in adjacent normal tissues was found mainly in the plasma membrane. The positive expression ratio of CLDN6 in adjacent tissues was 75.70% (81/107) and that in colon cancer tissues was only 26.16% (28/107) (Table 1). Besides, the expression of CLDN6 in cancer tissues was negatively associated with lymph node metastasis ( Table 2). We also detected the expression of CLDN6 in the colon cancer cell line SW1116. Both mRNA (Fig. 1b, c) and protein levels (Fig. 1d, e) of CLDN6 were significantly lower in SW1116 cells than that in NCM460 cells. Collectively, these results suggested that CLDN6 was lowly expressed in colon cancer and it may have an important role in the progression of colon cancer.

CLDN6 inhibits proliferation, migration and invasion of SW1116 cells
To better understand the function of CLDN6, we stably overexpressed CLDN6 in SW1116 cells. RT-PCR and WB were performed to measure the mRNA and protein levels of CLDN6 (Fig. 2a-c). As shown in Fig. 2d, e, the colony formation ability of SW1116/CLDN6 cells was significantly lower than that of SW1116/vec cells. Using CCK8 assay, we found that SW1116/CLDN6 cells showed lower cell viability compared with SW1116/vec cells (Fig. 2f). To evaluate whether CLDN6 inhibited cell proliferation in vivo, we injected subcutaneously tumor cell suspensions in nude mice and observed the situation of tumor growth. All six mice injected with SW1116/vec cells formed a transplanted tumor, as in the SW1116/CLDN6 group only five mice formed a transplanted tumor (Fig. 2g). Tranplanted tumors formed by SW1116/CLDN6 cells had a lighter weight (Fig. 2h) and slower growth rate (Fig. 2i) than that generated by SW1116/vec cells. IHC assay also verified the higher expression of CLDN6 in the SW1116/CLDN6 xenografts (Fig. 2j). These results suggested that CLDN6 inhibited the proliferation of SW1116 cells in vitro and in vivo. We also found that the cell morphology changed from fusiform to polygon after CLDN6 overexpression (Fig. 3a). Next, we performed a wound-healing assay to evaluate the effect of CLDN6 on cell migration. The results showed that the SW1116/CLDN6 cells had a significantly decreased wound-healing ability compared with SW1116/vec cells (Fig. 3b, c). SW1116/CLDN6 cells elicited a significant reduction of invading cells compared with SW1116/vec cells detected by transwell chamber assay (Fig. 3d, e). Taken together, these results indicated that CLDN6 suppressed the migration and invasion abilities of SW1116 cells.

CLDN6 suppresses the migration and invasion abilities of SW1116 cells via activating the TYK2/ STAT3 pathway
Our previous studies have shown that MCF-7/CLDN6 cells have significantly changed genes enriched in the TYK2/ STAT3 pathway. We next accessed the effect of CLDN6 on the TYK2/STAT3 pathway in SW1116 cells. As shown in Fig. 4a and b, p-TYK2 and p-STAT3 expressed significantly higher in SW1116/CLDN6 cells, indicating that CLDN6 activated the TYK2/STAT3 pathway.
To demonstrate whether CLDN6 exerted its function on cell migration and invasion abilities via the TYK2/STAT3 pathway, SW1116/CLDN6 cells were treated with AG490, an inhibitor of the TYK2/STAT3 pathway. After treating SW1116/CLDN6 cells with different concentrations of AG490, the optimal concentration was found to be 50 μM (Fig. 4c, d). Besides, SW1116/CLDN6 cells treated with AG490 for 48 h showed a significantly lower p-STAT3 expression (Fig. 4e, f). Thus, we used the most appropriate concentration and time of AG490 to process the following research. AG490 treatment returned the SW1116/CLDN6 cells to the original fusiform (Fig. 4g). The wound-healing assay showed that cells treated with AG490 had a significantly increased wound healing ability compared with the 1 3 control (Fig. 4h, i), indicating AG490 reversed the inhibition of migration by CLDN6. The invasion assay conducted using matrigel-coated transwell chambers obtained similar results (Fig. 4j, k). Collectively, these data indicated that CLDN6 might suppress the migration and invasion abilities of SW1116 cells via activating the TYK2/ STAT3 pathway.

Discussion
Dysregulated CLDN6 plays an oncogenic or tumor-suppressive role depending on target tissues or cell types. Studies have shown that CLDN6 was upregulated and promoted tumor progression in gastric cancer [27], non-small-cell lung cancer [28], atypical teratoid/rhabdoid tumors [29], and hepatocellular carcinoma [30], which made CLDN6 a possible therapeutic target for these tumors. In the present study, we found that CLDN6 was lowly expressed in colon cancer, and this was consistent with our previous study of low expression of CLDN6 in breast cancer [26]. In breast cancer, the reduction of CLDN6 mRNA is associated with the methylation of CLDN6's promoter [31,32]. The underlying mechanism of CLDN6 expression in colon cancer should be clarified in the future.
To evaluate the role of CLDN6, we overexpressed CLDN6 in SW1116 cells and examined the effect of CLDN6 on the proliferation of SW1116 cells by in vitro and in vivo experiments. The in vitro experiments showed that CLDN6 inhibited the colony formation ability and cell viability of SW1116 cells, suggesting that CLDN6 inhibited colon cancer cell proliferation. The in vivo experiments showed that the volume and weight of transplanted tumors in the SW1116/CLDN6 group were significantly lower than those in the SW1116/vec group, suggesting that CLDN6 inhibited the growth of transplanted tumors in nude mice. We noted that all six mice in the SW1116/vec group had tumors, while only five mice in the SW1116/CLDN6 group had tumors, which we speculate may be related to individual differences in nude mice or the manipulation of the cell suspension when it was inoculated.
The JAK/STAT pathway is considered to be the core pathway of tumors and directly contributes to all hallmarks of tumors [33,34]. Previous studies have mostly focused on the relationship between JAK1-3 and tumors, and as research has progressed, the impact of TYK2 on tumors has received increasing attention, especially its relationship with tumor metastasis [35,36]. Abundant evidence indicates that TYK2/ STAT3 pathway is persistently activated in several cancers, with a crucial position in tumor onset and progression [37,38]. Notably, recent studies have shown that a high level of CLDN9 or CLDN17 enhanced the metastatic potential of Fig. 3 Effects of CLDN6 on the migration and invasion abilities of SW1116 cells. a Comparison of cell morphology before and after CLDN6 overexpression. b, c The migration ability of SW1116/vec and SW1116/ CLDN6 cells was assessed by wound-healing assay. d, e The invasion ability of SW1116/ vec and SW1116/CLDN6 cells was assessed by transwell assay. *P < 0.05 hepatocytes via activating the TYK2/STAT3 pathway [39,40]. Consistent with this, in colon cancer cells, we found that the TYK2/STAT3 pathway was activated following CLDN6 overexpression. However, the activated TYK2/STAT3 pathway played an important role in CLDN6-mediated inhibition of proliferation, migration, and invasion, contrary to its role in CLDN9 or 17-mediated promoting the metastatic potential of hepatocellular carcinoma. Similar to our results, several studies have demonstrated the anticancer effect of STAT3 or p-STAT3. For example, Bekki et al. [41] found that in undifferentiated pleomorphic sarcoma, p-STAT3 expression was lower in tumors than in normal tissues, and positivity for p-STAT3 was significantly correlated with a better prognosis. Wu et al. [42] reported that high STAT3 expression may predict a better overall outcome for breast cancer patients. It's speculated that different cells or tissues, and different cellular states or microenvironments may affect the role of the TYK2/STAT3 pathway.
There are still certain limitations in this study. First, although we have observed that the expression of CLDN6 in colon cancer tissue was negatively related to colon cancer lymph node metastasis, we have only verified it in vitro, and in vivo experiments are needed to verify our results. Second, we found that CLDN6 overexpression activated the TYK2/ STAT3 pathway, and we used AG490 to inhibit the activation of the signaling and detect cell migration, and invasion. However, although AG490 was considered as a pan-JAK family inhibitor in some studies [43][44][45], it has also been reported to be a specific inhibitor of JAK2 [46][47][48]. So we can't exclude its inhibitory effect on the JAK2/STAT3 pathway. Finally, the mechanism by which CLDN6 increased the expression level of p-TYK2 was not clear. CLDNs play a decisive role in maintaining the integrity of the barrier function of epithelial cells. After the redistribution or expression of CLDNs change, the tight junctions between cells are destroyed and the barrier function of epithelial cells is impaired, which in turn changes the permeability between epithelial cells and mediates cellular inflammation. The JAK/STAT pathway is regulated by a variety of cytokines including distinct interleukin (ILs) and interferons (IFNs), which mediate the pathway activation by binding to corresponding receptors. It is reported that the loss of CLDN3 promotes the IL-6/STAT3 signaling [14]. Besides, IL-12 and IL-23 specifically activate the TYK2/STAT3 pathway [49]. Whether CLDN6 overexpression regulates the TYK2/ STAT3 pathway by affecting cytokines including IL-12 or IL-23 deserves further exploration. Besides, the PDZ binding motif of CLDN6 may play an important role in activating the TYK2/STAT3 pathway. Importantly, a recent study showed that CLDN6 directly activated the SFK/PI3K/AKT signaling depending on ECL2 and Y196/200 to transmit cell adhesion signals to the nucleus and regulate gene expression [20]. To expand upon the results of the present study, our future investigations will attempt to elucidate the mechanism by which CLDN6 activates the TYK2/STAT3 pathway.

Conclusion
CLDN6 expression was low in colon cancer and negatively correlated with lymph node metastasis in patients. CLDN6 inhibited the proliferation of SW1116 cells in vitro and in vivo. CLDN6 may suppress the migration and invasion abilities of SW1116 cells by activating the TYK2/STAT3 pathway. Our finding suggested that CLDN6 acted as a tumor suppressor and could be a potential biomarker for colon cancer progression.
Author contributions The first draft of the manuscript was written by HQ. Cell and animal experiments, data analysis were performed by MW. Molecular biology experiments were performed by MW. Immunohistochemistry experiments were performed by YL. Study design, funding acquisition, and supervision were completed by CQ.

Data availability
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Conflict of interest
The authors have no relevant financial or non-financial interests to disclose.
Ethical approval Tumor tissues were obtained from the Eastern Division of the First Hospital of Jilin University. This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethics Committee of Jilin University and informed consent was obtained from all patients. All animal experiments and their care were approved by the Experimental Animal Ethical Committee of Jilin University and were carried out following relevant institution guidelines. Fig. 4 CLDN6 suppresses the migration and invasion abilities of SW1116 cells by activating the TYK2/STAT3 pathway. a, b WB analysis of proteins in the TYK2/STAT3 pathway of SW1116/vec and SW1116/CLDN6 cells. c, d SW1116/CLDN6 cells were treated with AG490 at indicated concentrations for 24 h to find the optimum treatment dose. e, f SW1116/CLDN6 cells were treated with 50 μM AG490 for indicated hours to find the optimum treatment time. g The morphology of SW1116/CLDN6 cells before and after AG490 treatment. h, i The effect of AG490 on the migration ability of SW1116/ CLDN6 cells by wound-healing assay. j, k The effect of AG490 on the invasion ability of SW1116/CLDN6 cells by transwell assay. *P < 0.05 ◂