Aberrant Claudin-6–Adhesion Signaling Promotes Endometrial Cancer Progression via Estrogen Receptor α

Claudins (CLDN) are the structural and functional backbone of tight junctions, the apical-most apparatus of apical junctional complexes in vertebrate epithelial cell sheets (1–4). The CLDN family consists of more than 20 members in mammals, and shows diverse expression patterns in tissue- and cell-type specific fashions. CLDNs also frequently exhibit aberrant expression and/or localization in a wide variety of cancer tissues depending on each CLDN subtype (5–8). CLDNs are tetraspan membrane proteins with a short cytoplasmic N-terminus, two extracellular loops (EC1 and EC2), and a C-terminal cytoplasmic domain. The CLDN–EC1 contributes to the paracellular barrier function to control selective transport of ions and substances. On the other hand, the CLDN–EC2 is involved not only in the binding of Clostridium perfringens enterotoxin (CPE), but also in trans-interaction between the plasma membranes of adjacent cells. Moreover, the C-terminal cytoplasmic domain of CLDNs is regarded to possess signaling properties, but the molecular mechanism remains poorly defined (9).

Among the CLDN members, CLDN6 displays a unique expression profile in normal mammalian cells. The CLDN6 expression is strongly induced in trophectoderm (10) and during differentiation of embryonic stem cells and F9 stem cells into extraembryonic visceral endoderm cells (11–15). It is also expressed in a broad range of embryonal epithelial cells such as epiblasts, hypoblasts, and definitive endoderm, as well as epithelia of organs including the lung, gut, pancreas, and kidney (16). On the other hand, CLDN6 protein is not detectable in any cell and tissue types of adult mouse and human organs (17).

CLDN6 is also known to be highly expressed in germ cell tumors, including seminoma, embryonal carcinoma, and yolk sac tumor, as well as in certain cases of gastric adenocarcinoma, lung adenocarcinoma, ovarian adenocarcinoma, and endometrial carcinoma (17–20). We have recently shown that aberrant CLDN6 expression is detected in approximately 6% of endometrial cancer subjects and significantly associated with surgical stages III/IV, histologic type, histologic grade 3, lymphovascular space involvement, lymph node metastasis, and distant metastasis (21). In addition, we have demonstrated that the high CLDN6 expression in endometrial cancer represents an independent prognostic factor, and the 5-year survival rate is about 30%, which is one third of that in the low expression group (21). However, the mechanism by which abnormal CLDN6 expression influences endometrial cancer progression is still unknown.

We previously identified that CLDN6 recruits and activates Src-family kinases (SFK) in the EC2-dependent and Y196/200-dependent manners, and the CLDN6/SFK/PI3K/AKT axis stimulates the retinoic acid receptor γ (RARγ) and estrogen receptor α (ERα) activity (22). Taken collectively with the concept that ERα serves as a master transcription factor in endometrial cancers (23), we hypothesized that the CLDN6 signaling modulates the malignant phenotypes of endometrial cancer cells via ERα. Here, we report that the CLDN6/SFK/PI3K axis propagates AKT and SGK (serum- and glucocorticoid-regulated kinase) and targets ERαS518, leading to stimulation of the ERα activity and malignant behaviors in endometrial cancer cells.

The antibodies used in this study are listed in Supplementary Table S1. A rabbit pAb against CLDN6 was generated in cooperation with Immuno-Biological Laboratories as described previously (14). Rat mAbs against CLDN6 were established as reported previously (21).

The Ishikawa cell lines #1 and #2 were obtained from Dr. Yamada (Wakayama Medical University) and Kasumigaura Medical Center, respectively. The HEC-1A and HEC-1B cell lines (24) were obtained from National Institute of Biomedical Innovation, Health and Nutrition (Japan). The KLE and RL-95 cell lines were obtained from ATCC, and the HHUA cell line was received from RIKEN. F9:Cldn6 was established previously (15). Cells were grown in RPMI1640 (Ishikawa), McCoy's 5A (HEC-1A), DMEM (HEK293T and F9:Cldn6), or DMEM with Nutrient Mixture F-12 (DMEM/F-12; HEC-1B, KLE, HHUA, and RL-95), with 10% FBS (Sigma-Aldrich), and 1% penicillin–streptomycin mixture (ChemScence). Ishikawa and HEC-1A cells were treated with 1 μmol/L of C-CPE, 10 μmol/L of PP2 (Calbiochem), 10 μmol/L of LY294002 (Cell Signaling Technology), 10 μmol/L of AKT inhibitor VIII (funakoshi), or 0.1 nmol/L of SGK-1 inhibitor (Santa Cruz Technology) 1 or 2 days after plating. For preparation of charcoal-treated FBS, 500 mL of FBS was treated with 0.5 g of charcoal, dextran coated (Sigma) overnight at 4°C followed by filtration using 0.22 μm cellulose acetate filter membranes. C-CPE production and purification were performed as described previously by using Escherichia coli BL21 and the expression vector pET16b coding C-CPE194–319 (25). Mycoplasma testing was not performed. Cells two or three passages after thawing were used in the described experiments.

To establish the Ishikawa:ESR1^−/− cell line, a pair of transcription activator-like effector nucleases (TALEN) targeting the second exon of ESR1 gene were designed by TALEN Targeter 2.0 software (https://tale-nt.cac.cornell.edu/node/add/talen; ref. 26). The expression vectors of the TALENs were cloned by using Platinum TALEN Kit (27). The plasmids were transiently transfected by Polyethylenimine Max (PEI Max, Cosmo Bio). Next, 24 to 48 hours after transfection, the cells were exposed to 100 μg/mL of hygromycin for positive selection, followed by limiting dilution and genotyping by PCR-based restriction fragment length polymorphism (RFLP; ref. 28) utilizing the endogenous Stu I recognition site. Knockout of ESR1 genes was verified by DNA sequence after TA-cloning of genomic PCR products.

The protein coding regions of human CLDN6 and/or ESR1 were cloned into the BamHI/NotI site of the CSII-EF-MCS-IRES2-Venus (RIKEN, RDB04384) plasmid. Hemagglutinin (HA) tag was added by PCR with tailed primer. Expression vectors of mutant genes (CLDN6Y196A, CLDN6Y200A, ESR1ΔC, and ESR1S518A) were established by a standard site-directed mutagenesis protocol using KOD -Plus- Mutagenesis Kit (TOYOBO) following the provider's protocol.

For transient expression of HA-ESR1ΔC and HA-ESR1S518A genes, Ishikawa:ESR1^−/− cells were transfected with 10 μg of the indicated vectors using 30 μg of Polyethylenimine Max (PEI Max; Polysciences) 8 hours after passage.

Ishikawa:CLDN6, Ishikawa:CLDN6Y196A, Ishikawa:CLDN6Y200A, Ishikawa:ESR1^−/−:ESR1, Ishikawa:ESR1^−/−:ESR1ΔC, Ishikawa:ESR1^−/−:CLDN6:ESR1S518A, HEC-1A:CLDN6, HEC-1A:ESR1, and HEC-1A:ESR1:CLDN6 cell lines were established by lentiviral transfection. First, lentiviral vectors (CLDN6, CLDN6Y196A, CLDN6Y200A, ESR1, ESR1ΔC, and ESR1S518A) were generated by transfecting HEK293T cells with 10 μg of the CSII plasmids containing the target genes, 5 μg of packaging plasmids psPAX2 (Addgene, #12260) and pCMV-VSV-G (Addgene, #8454) using PEI Max. Culture media containing recombinant lentiviruses were collected 72 hours after transfection. The lentiviral vectors were added to the cell culture medium of Ishikawa, Ishikawa:ESR1^−/−, or HEC-1A cell lines after filtration. More than 48 hours after transfection, the cells were used for further analysis. Ishikawa:CLDN6 cell lines were single cell-cloned by limiting dilution with 96-well culture plates.

Spheroids were obtained by a hanging drop cell culture model. First, equal number of either Ishikawa and Ishikawa:CLDN6 or Ishikawa:ESR1^−/− and Ishikawa:ESR1^−/−:CLDN6 cells were mixed and well singularized by pipetting. Then 3 × 10³ cells in 30 μL are placed on the inner side of a lid of 10 cm cell-culture dish, and the lid was placed on culture dish filled with 20 mL of PBS. The spheroids were collected after 2 days, and grown on coverslips coated by Cellmatrix Type I-A (Nitta gelatin) for additional two days, followed by phase-contrast and immunofluorescence analyses.

Total cell extracts were collected by using CellLyticTM MT Cell Lysis Reagent (Sigma), and were subsequently sonicated with three or four bursts of 5 to 10 seconds. Immunoprecipitation was performed using an Immunoprecipitation Kit (Protein G; Sigma), following the manufacturer's protocol. One to two micrograms of ChromPure Mouse (Fig. 2E and F) and Rat (Figs. 2C and D, and 5B; Jackson Immunoresearch Laboratories) were used as negative controls. Whole cell lysates or the immunopreciptated samples were mixed with sample loading buffer containing 2-mercaptoethanol and incubated for 10 minutes at 95°C. They were resolved by one-dimensional SDS-PAGE and electrophoretically transferred onto a polyvinylidene difluoride membrane. The membranes were blocked with TBS containing 4% skim milk for 30 minutes. After rinsing in TBS containing 0.1% Tween 20, the membranes were incubated with a primary antibody solution for 1 hour at room temperature or overnight at 4°C, followed by incubation with HRP-conjugated secondary antibodies. They were rinsed again and exposed to EzWestLumi One (ATTO). After rinsing with 10% H₂O₂ to inactivate HRP, each membrane was hybridized with HRP-conjugated anti-β-actin antibody as loading controls. Each signal was quantified by ImageJ software (Wayne Rasband National Institutes of Health) and divided by the corresponding actin levels.

Cells were grown on coverslips coated by Cellmatrix Type I-A (Nitta gelatin). The samples were fixed in 1% paraformaldehyde and 0.1% Triton-X for 10 minutes at room temperature. After washing with PBS, they were preincubated in PBS containing 5% skimmed milk. They were subsequently incubated overnight at 4°C with primary antibodies in PBS, then rinsed again with PBS, followed by a reaction for 1 hour at room temperature with appropriate secondary antibodies. All samples were examined using a laser-scanning confocal microscope (FV1000, Olympus). Photographs were processed with Photoshop CC (Adobe) and ImageJ software (Wayne Rasband National Institutes of Health).

For IHC staining, uterine endometrial carcinoma and xenograft tumor tissues were obtained, and the 10% formalin-fixed and paraffin-embedded (FFPE) tissue blocks were sliced into 5-μmol/L-thick sections, then deparaffinized with xylene and rehydrated using a graduated series of ethanol. The sections were immersed in 0.3% hydrogen peroxide in methanol for 20 minutes at room temperature to block endogenous peroxidase activity. Antigen retrieval was performed by incubating the sections in boiling citric acid buffer (pH 6.0) in a microwave. After blocking with 5% skimmed milk at room temperature for 30 minutes, the sections were incubated overnight at 4°C with the primary antibodies. Histofine SAB-PO Kit for rabbit (Nichirei) or VECTASTAIN Elite ABC HRP Kit for rat (Vector Laboratories) was used for 3′,3′-diaminobenzidine (DAB) staining. The study was approved by the ethics committee of Fukushima Medical University Hospital (FMUH).

Cell proliferation index was evaluated by incorporation of bromodeoxy uridine (5-Bromo-2-DeoxyUridine, BrdU; Sigma). Cells were exposed to BrdU for 5 minutes after passage. The specimens were fixed with 4% paraformaldehyde and 0.1% Triton-X, followed by immunostaining with anti-BrdU antibody (BD) and its standard protocol.

For evaluating cell migration, wound areas were generated by scratching with disposable 1,000 μL pipette tips 24 to 48 hours after passage. Culture media were changed daily. Photographs of the wound areas were taken at the same locations 0 and 2 days after scratching, using a phase-contrast microscope. Wound healing was calculated as the percentage of the remaining cell-free area compared with the initial wound area using ImageJ software. In situ Cell Death Detection Kit (Roche) was used for evaluation of cell apoptosis.

Xenograft studies were performed in 8-week-old NOD/ShiJic-scid female mice (CLEA-Japan). A total of 1×10⁷ cells were subcutaneously injected into the back of anesthetized mice. Then, 28 or 30 days after injection, the mice were ethically sacrificed. All animal experiments conformed to the National Health Guide for the Care and Use of Laboratory Animals, and were approved by the Animal Committee at Fukushima Medical University.

Total RNA was isolated from cells using TRizol RNA Isolation Reagents (Thermo Fisher Scientific). To collect total RNAs from surgical specimens, unstained sections obtained from FFPE blocks were deparaffinized by rinsing in 100% xylene, 100% ethanol, and PBS. The sections were incubated in 5U proteinase K in PBS at 55°C for 2 hours and PBS at 85°C for 15 minutes. Then, the tissue was collected in a 1.5 mL tube by a razor and the total RNA was extracted by TRizol RNA Isolation Reagents (Thermo Fisher Scientific).

For RT-qPCR, reverse transcription was performed using Primescript II RT Kit (Clontech) and target genes were quantified by THUNDERBIRD SYBR qPCR Mix (TOYOBO) and Step One Real-Time PCR System (Applied Biosystems) using the primers listed in Supplementary Table S2. The expression levels of the target genes were divided by the corresponding GAPDH signal intensity.

RNA sequencing and mapping were performed by TaKaRa Bio Inc. For mapping, the index trimmed single-end 100 bp reads were aligned to the human reference genome (GRCh38 v90) to generate bam files. The mapped bam files were imported to SeqMonk software (Babraham Bioinformatics; https://www.bioinformatics.babraham.ac.uk/projects/seqmonk/) as single-ended RNA-seq data. Then they were quantitated by using the default RNA-seq quantitation pipeline. P value was calculated from the intensity difference and representative genes of showing significant change (P < 0.05) are listed in Fig. 6A. Raw data were uploaded to Gene Expression Omnibus (https://www.ncbi.nlm.nih.gov/geo/) as GSE161384.

Statistical significance for cell proliferation, migration, and xenograft studies was analyzed by unpaired two-tailed Student t test.

We first determined, by Western blot analysis, the expression of CLDN6 and ERα in representative human endometrial carcinoma cell lines, such as Ishikawa, HHUA, HEC-1A, HEC-1B, KLE, and RL95. As shown in Supplementary Fig. S1A, CLDN6 protein was not detected in these cell lines. On the other hand, the expression of ERα protein appeared in Ishikawa #1 cells, but was hardly detected in other endometrial carcinoma cell lines.

Because no human endometrial carcinoma cell lines expressed endogenous CLDN6 protein as far as we examined, we generated, using the lentiviral vector system, Ishikawa #1 cells expressing CLDN6 (Ishikawa:CLDN6; Fig. 1A; Supplementary Fig. S1B). CLDN6 was concentrated along the cell borders in Ishikawa:CLDN6 cells, indicating that CLDN6 acts as a cell adhesion molecule (Fig. 1B). It was reasonable that expression levels of CLDN6 in Ishikawa:CLDN6 cells were higher than those in the CLDN6-high cases of endometrial cancer tissues, because in the latter of which large number of interstitial cells were contained (Supplementary Figs. S1C and S1D). BrdU assay revealed that cellular proliferation was significantly increased in Ishikawa:CLDN6 cells compared with parental Ishikawa cells (Fig. 1C and D). In contrast, on the TUNEL assay, few apoptotic cells were observed in both cell lines (Supplementary Fig. S2). Moreover, wound healing assay demonstrated that cell migration in Ishikawa:CLDN6 cells was significantly accelerated compared with that in Ishikawa cells (Fig. 1E and F).

We then validated whether the high CLDN6 expression also promoted malignant phenotypes of human endometrial carcinoma cells in vivo. Four weeks after inoculation in SCID mice, the tumor growth of Ishikawa:CLDN6 xenografts were significantly increased compared with that of Ishikawa xenografts (Fig. 1G and H). Neither lymph node nor distant metastasis was grossly evident in these xenografts. Microscopically, Ishikawa:CLDN6 xenografts were equivalent to Grade 3 endometrial carcinomas that were rich in solid components (Fig. 1I). Furthermore, intratumor heterogeneity of CLDN6 expression was observed in Ishikawa:CLDN6 xenograft tissues as in the high CLDN6 expression cases of endometrial cancer subjects (21). It is also noteworthy that invasion into the fibrous capsule around the tumor was prominent in Ishikawa:CLDN6 xenografts but hardly in Ishikawa ones.

We next verified the involvement of CLDN6-EC2 and CLDN6-Y196/200 in activation of SFKs and formation of the CLDN6/pSFK complex in human endometrial carcinoma cells. Double immunofluorescence staining showed that pSFK appeared to be concentrated to cell boundaries together with CLDN6 in Ishikawa:CLDN6 cells (Fig. 2A). In contrast, pSFK signal was scarcely detected on cell borders in parental Ishikawa cells. When Ishikawa:CLDN6 cells were exposed to C-terminal half of CPE (C-CPE), which binds to the EC2 of CLDN6 and excludes CLDN6 from cell membranes without alteration in its total protein levels (16, 19), the pSFK immunoreactivity was markedly reduced. On Western blot analysis, the levels of pSFK were elevated in Ishikawa:CLDN6 cells compared with Ishikawa cells, and decreased in both Ishikawa:CLDN6Y196A and Ishikawa:CLDN6Y200A cells (Fig. 2B). Immunoprecipitation assay revealed that CLDN6 was associated with pSFK in Ishikawa:CLDN6 cells, and the CLDN6/pSFK complex was diminished in Ishikawa:CLDN6 cells on C-CPE treatment as well as in Ishikawa:CLDN6Y196A and Ishikawa:CLDN6Y200A cells (Fig. 2C and D).

We also demonstrated that CLDN6 was highly tyrosine-phosphorylated in Ishikawa:CLDN6 cells, and the phospho-tyrosine levels were suppressed by C-CPE exposure and in both Ishikawa:CLDN6Y196A and Ishikawa:CLDN6Y200A cells (Fig. 2E and F). In addition, the promoted cell proliferation and migration in Ishikawa:CLDN6 cells were significantly reversed by C-CPE treatment (Fig. 1D and F). Moreover, the CLDN6-enhanced cell proliferation was prevented in Ishikawa:CLDN6Y196A or Ishikawa:CLDN6Y200A cells (Fig. 2G). Taken collectively, these results indicated that the CLDN6 signaling activated SFKs and accelerated endometrial cancer progression in the EC2- and Y196/200-dependent manners.

We subsequently validated the involvement of PI3K and the two major downstream cascades AKT and SGK (serum- and glucocorticoid-regulated kinase), which shares the high degree of homology and the same consensus phosphorylation motif (29), in the CLDN6/SFK signaling. To achieve this goal, we used the respective protein kinase inhibitors LY294001, AKT inhibitor VIII and SGK1 inhibitor. As expected, the enhanced cell proliferation in Ishikawa:CLDN6 cells was prevented by these inhibitors and the SFK inhibitor PP2 (Supplementary Fig. S3A). Besides, in the presence of the inhibitors, CLDN6 less efficiently accelerated cell migration of Ishikawa cells (Supplementary Fig. S3B).

To evaluate whether the CLDN6-adhesion signaling stimulates the malignant behavior of endometrial carcinoma cells via ERα, we then generated both Ishikawa:ESR1^−/− and Ishikawa:ESR1^−/−:CLDN6 cells, and compared their phenotypes. We designed the TALEN expression vector that is expected to excise the flanked DNA in exon 2 of ESR1 genes (Fig. 3A). Knockout of ESR1 genes was verified by DNA sequence (Fig. 3A), and the lack of ERα protein in Ishikawa:ESR1^−/− and Ishikawa:ESR1^−/−:CLDN6 cells was confirmed by Western blot analysis and immunostaining (Fig. 3B and C). Note that CLDN6 did not increase cell proliferation (Fig. 3D and E) or migration capacity (Fig. 3F and G) in Ishikawa cells in the absence of ERα.

We also used HEC-1A cells, in which neither CLDN6 nor ERα were expressed, and established cell lines expressing either CLDN6, ERα, or both together (Fig. 4A). Expression and subcellular localization of both CLDN6 and ERα protein were confirmed by Western blot analysis and immunostaining (Fig. 4B and C). Cell growth was significantly elevated in HEC-1A:ESR1:CLDN6 cells but not in HEC-1A:CLDN6 or HEC-1A:ESR1 cells compared with parental HEC-1A cells (Fig. 4D). Cell migration was also significantly increased in HEC-1A:ESR1:CLDN6 cells compared with HEC-1A cells, and was raised in HEC-1A:CLDN6 and HEC-1A:ESR1 cells less efficiently than in HEC-1A:ESR1:CLDN6 cells (Fig. 4E). In addition, exposure of HEC-1A:ESR1:CLDN6 cells to C-CPE significantly reversed the increase in cell proliferation and migration (Fig. 4D and E), again indicating the critical role of the EC2 in the CLDN6 signaling. Furthermore, the tumor growth of HEC-1A:ESR1:CLDN6 xenografts was significantly increased compared with those of HEC-1A, HEC-1A:CLDN6, or HEC-1A:ESR1 xenografts (Fig. 4F and G). Of note, extensive invasion into the tumor capsule appeared in HEC-1A:ESR1:CLDN6 xenografts, but neither in HEC-1A, HEC-1A:CLDN6, nor HEC-1A:ESR1 ones (Fig. 4H). Taken all together, these results strongly suggested that the CLDN6-adhesion signaling links to ERα in endometrial cancer cells, resulting in tumor progression.

Interestingly, AKT and SGK1 were associated with transiently introduced ERα, but not with ERαΔC, in Ishikawa:ESR1^−/− cells (Fig. 5A and B), indicating that both kinases target either the LBD/AF2 domain (E region) or F region of ERα. We next determined whether the CLDN6 signaling directs to ERαS518 and ligand (estradiol)-independently stimulated the ERα activity in endometrial cancer cells, as in the human breast cancer cell line MCF-7 (19). To this end, we generated Ishikawa:CLDN6:ESR1^−/−:ESR1-wt (wild-type) and Ishikawa:CLDN6:ESR1^−/−:ESR1S518A cells, in the latter of which ERαS518 was substituted for an alanine residue, and performed rescue experiments. These cells were also grown in phenol red-free medium with charcoal-treated FBS to exclude fat-soluble ligands. It should be noteworthy that the level of ERα protein in Ishikawa:CLDN6:ESR1^−/−:ESR1-wt cells was similar to that in Ishikawa:CLDN6:ESR1^−/−:ESR1S518A cells (Fig. 5C), and that CLDN6-triggered SFK and AKT activation was not basically influenced by the S518A mutation (Fig. 5D). The transcript levels of the four ER target genes (BCL2, CCND1, MYC, and VEGFA; ref. 30) were significantly higher in Ishikawa:CLDN6 cells than in Ishikawa cells (Fig. 5E). More importantly, the expression levels of these target genes were significantly reduced in Ishikawa:CLDN6:ESR1^−/−:ESR1S518A cells compared with those in Ishikawa:CLDN6:ESR1^−/−:ESR1-wt cells, and similar to those in Ishikawa:ESR1^−/− cells (Fig. 5F). Furthermore, cell migration was significantly diminished in Ishikawa:CLDN6:ESR1^−/−:ESR1S518 cells compared with those in Ishikawa:CLDN6:ESR1^−/−:ESR1-wt cells (Fig. 5G). In addition, cell proliferation was decreased in in HEC-1A:CLDN6:ESR1S518A cells compared with those in HEC-1A:CLDN6:ESR1-wt cells (Fig. 5H). Hence, the CLDN6-adhesion signaling directs to ERαS518 for promoting the ERα activity and malignant phenotypes in endometrial cancer cells.

Because prominent invasion was observed in Ishikawa:CLDN6 and HEC-1A:ESR1:CLDN6 xenografts, we subsequently verified the involvement of CLDN6–ERα signaling in collective migration of endometrial cancer cells. When spheroids composed of Ishikawa and Ishikawa:CLDN6 cells were grown on collagen-coated dish, Venus-expressing cells were striking in the migrating leading front (Fig. 6A). In addition, the CLDN6-immunoreactive signal was evident along lateral sides of adjacent leader cells, and to some extent detected in free surfaces of the leading edge, in which CLDN6 was at least in part colocalized with the actin belt (Fig. 6B). In contrast, such migration was not obvious in co-culture of Ishikawa:ESR1^−/− and Ishikawa:ESR1^−/−:CLDN6 cells. Moreover, the proportion of CLDN6-positive cells at the front edge were significantly increased in co-culture of Ishikawa/Ishikawa:CLDN6 cells compared with that in Ishikawa:ESR1^−/−/Ishikawa:ESR1^−/−:CLDN6 cells (Fig. 6C).

To identify downstream molecules that expression is altered by the CLDN6 signaling, we next compared, using RNA sequencing, the transcriptome in Ishikawa:CLDN6 cells with that in Ishikawa cells. Following the analysis, we listed up the top 50 genes that expression was significantly down- or upregulated in both of two distinct Ishikawa:CLDN6 cell lines compared with parental Ishikawa cells (Fig. 7A). We then by RT-qPCR compared the expression of 31 genes, which products are known to be associated with malignant phenotypes, in Ishikawa, Ishikawa:CLDN6, Ishikawa:ESR1^−/−, and Ishikawa:ESR1^−/−:CLDN6 cells (Fig. 7B). Among CLDN6-acitivating 19 genes, the expression of 9 genes (cluster #1) appeared to be induced via ERα, whereas 6 genes (cluster #2) were upregulated in an ERα-independent manner. In addition, the expression of 2 genes (cluster #3) were activated partially through ERα, and 2 genes (cluster #4) were likely up-regulated by both ERα-independent CLDN6 and ERα-dependent non-CLDN6 signaling. Thus, the CLDN6-activated genes can be classified into at least four groups with distinct ERα-dependence. It should also be noted that, out of CLDN6-suppressing 12 genes determined, 7 genes (cluster #5) seemed to be regulated via ERα-independent CLDN6 and ERα-dependent non-CLDN6 signaling.

In this study, we demonstrated that CLDN6 accelerated endometrial cancer progression in vitro and in vivo. This was obvious because introduction of the human CLDN6 gene was enough to promote cell proliferation and migration in two distinct endometrial cancer cell lines Ishikawa and HEC-1A:ESR1. In addition, overexpression of CLDN6 in both cell lines led to enhanced tumor growth and invasion into the fibrous capsule in xenografts. Thus, we established the biological relevance of the high CLDN6 expression in endometrial cancer. This conclusion is in good agreement with our recent work showing that aberrant CLDN6 expression in endometrial cancer tissues is significantly associated with several clinicopathological variables, including surgical stage III/IV, histological type, histological grade 3, lymphovascular space involvement, lymph node metastasis, and distant metastasis, and represents an independent prognostic marker for overall survival (21).

Another finding of this study is that the EC2 and Y196/200 of CLDN6 are responsible for recruiting and activating SFKs in endometrial cancer cells, as well as promoting the malignant properties. This conclusion was drawn from the following results: (i) the pSFK levels were increased in Ishikawa:CLDN6 cells but not in Ishikawa:CLDN6Y196A or Ishikawa:CLDN6Y200A cells; (ii) colocalization of CLDN6 and pSFK along cell boundaries was evident in Ishikawa:CLDN6 cells, and diminished by C-CPE treatment; (iii) a CLDN6–pSFK complex was formed in Ishikawa:CLDN6 cells, and their association was decreased upon C-CPE exposure and in Ishikawa:CLDN6Y196A and Ishikawa:CLDN6Y200A cells; (iv) the increased cell growth and migration in both Ishikawa:CLDN6 and HEC-1A:ESR1:CLDN6 cells were abrogated upon C-CPE treatment; (v) the CLDN6-stimulated cell proliferation was not detected in Ishikawa:CLDN6Y196A or Ishikawa:CLDN6Y200A cells. We also demonstrated that SFKs in turn phosphorylated CLDN6 at both Y196 and Y200, and tyrosine-phosphorylation of CLDN6 was governed by the EC2 domain. We previously reported that similar reciprocal regulation between CLDN6 and SFKs is also observed in mouse F9 stem cells (22), further strengthening our conclusion. Moreover, using the respective protein kinase inhibitors, we revealed that the PI3K-dependent AKT and SGK cascades contributed to the CLDN6/SFK signaling in endometrial cancer progression.

The most important conclusion of this work is that the CLDN6/SFK/PI3K-dependent AKT and SGK signaling target ERα in endometrial cancer cells. This was apparent because CLDN6-accerelated cell growth and migration were hindered in Ishikawa:CLDN6:ESR1^−/− cells. Using HEC-1A expressing CLDN6 and/or ERα, it was confirmed that CLDN6–ERα signaling promotes endometrial cancer advancement in vitro and in vivo. Furthermore, AKT and SGK1 formed a complex with ERα in endometrial cancer cells, reinforcing the conclusion. In contrast, neither kinases were associated with ERαΔC, indicating that they do not directly target the known AKT substrate S167 (31–34) at least in endometrial cancer cells. Instead, our RT-qPCR analysis indicated that the CLDN6 signaling directs to S518 in ERα and ligand-independently activated various oncogenic target genes. We also revealed that ERα-S518 is responsible for the CLDN6-accelerated malignant behaviors in endometrial cancer cells. The pathobiologic relevance of the ERαS518 phosphorylation should be determined not only in endometrial cancer tissues, but also in other hormone-dependent tumors, such as ovarian cancer and breast cancer, in future experiments.

It is well known that mechanosensitive cell adhesion, especially cadherin-based adherens junctions, plays a fundamental role in collective cell migration (35–42); however, it has yet to be defined whether and how cell adhesion signaling contributes to collective cell behaviors. Using a co-culture system of either Ishikawa and Ishikawa:CLDN6 or Ishikawa:ESR1^−/− and Ishikawa:ESR1^−/−:CLDN6 cells, we in this study demonstrated that CLDN6–ERα signaling is required for collective migration of endometrial cancer cells, especially for collective cell behaviors in the leading front. This conclusion was further supported by our finding that massive invasion into tumor capsule was prominent in HEC-1A:ESR1:CLDN6 xenografts, but not in HEC-1A, HEC-1A:CLDN6 or HEC-1A:ESR1 ones. Thus, our results provided novel insight into regulation of collective cell migration by cell adhesion–transcription factor signaling.

Our RNA-seq analysis revealed that a variety of gene expression, including the SGK1 gene, was altered between Ishikawa and Ishikawa:CLDN6 cells. Among such genes, we focused on CLDN6-upregulated and downregulated genes that correspond to proto-oncogenes and tumor suppressor genes, respectively. On RT-qPCR and cluster analyses indicated that the CLDN6-activated genes were categorized into four groups with requirement of ERα. Out of CLDN6-upregulated 19 genes, the expression of 9 and 6 genes was induced in ERα-dependent and independent manners, respectively. Enhanced cell migration was observed in HEC-1A:CLDN6 cells without ERα expression, supporting the presence of not only ERα-dependent but also ERα-independent CLDN6 signaling. On the other hand, the majority of CLDN6-downregulated genes seemed to be controlled by both ERα-independent CLDN6 and ERα-dependent non-CLDN6 signaling. Interestingly, the novel AKT/SGK-consensus phosphorylation motif is conserved in 14 of 48 members of human nuclear receptors (22). Taken together, CLDN6 may also target these nuclear receptors and possibly other transcription factors to regulate the expression of certain genes. In fact, we previously reported that the CLDN6 signal targets RARγ in mouse F9 stem cells to initiate epithelial differentiation (22).

In summary, we here established that high expression of CLDN6 protein in endometrial cancer leads to more aggressive tumors. We also demonstrated that the CLDN6/SFK/PI3K-dependent AKT and SGK cascades direct to S518 in human ERα and stimulated its activity, resulting in progression of tumor behaviors in endometrial cancer. Therefore, in addition to the PI3K/AKT pathway, which is frequently altered in endometrial cancers (43–46), the CLDN6/SFK, SGK and ERαS518 may be promising therapeutic targets for endometrial cancer. It would also be interesting to determine whether a similar link between cell adhesion and nuclear receptor signaling regulates tumor progression in various types of cancers.

No disclosures were reported.

M. Kojima: Funding acquisition, validation, investigation, methodology, writing–original draft. K. Sugimoto: Conceptualization, supervision, funding acquisition, validation, investigation, methodology, writing–original draft, writing–review and editing. M. Kobayashi: Validation, investigation, methodology. N. Ichikawa-Tomikawa: Validation, investigation. K. Kashiwagi: Validation, investigation. T. Watanabe: Validation. S. Soeda: Validation. K. Fujimori: Validation. H. Chiba: Conceptualization, supervision, funding acquisition, validation, writing–original draft, project administration, writing–review and editing.

We thank M. Nishita for his technical advice; A. Hozumi and K. Watari for their technical assistance, and English Editing Service of the Medical Research Promotion Office, Fukushima Medical University for their assistance with the manuscript. This work was supported by JSPS KAKENHI (Grant Nos. 17K08699, 17K17978, and 17K17981), by AMED (Grant No. JP20lm0203010), and by the Uehara Memorial Foundation and the Takeda Science Foundation.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.