CLDN6 promotes tumor progression through the YAP1-snail1 axis in gastric cancer

Abstract

Claudin6 (CLDN6), a member of the tight junction family, is a molecule involved in intercellular adhesion, acting as a physical barrier that prevents solutes and water from freely passing through the extracellular space. CLDN6 has important biological functions, and its abnormal expression is associated with Hepatitis C infection. However, there is limited research regarding its role in gastric cancer. In this study, we found that the expression of CLDN6 mRNA and protein was upregulated in gastric cancer cell lines and tissues, which indicated poor prognosis. Both in vitro and in vivo experiments showed that abnormal CLDN6 expression was associated with enhanced proliferation and invasion abilities of gastric cancer. CLDN6 reduced the phosphorylation of LATS1/2 and YAP1 by interacting with LATS1/2 in the Hippo signaling pathway. Thus, CLDN6 affected the entry of YAP1 into the nucleus, causing changes in downstream target genes. Moreover, YAP1 interacted with snail1 to affect the process of EMT and enhanced the invasive ability of GC cells. Collectively, CLDN6 promoted the proliferation and invasive ability of gastric cancer by affecting YAP1 and YAP1-snail1 axis.

Fig. 1. CLDN6 is upregulated in gastric cancer.

a Screening process of differentially expressed genes from TCGA dataset (tumor samples [n = 32], normal samples [n = 32]). b Upregulated and downregulated genes in volcano plot. c Major pathways identified from DAVID fold enrichment analysis. d Expression profiles of different claudins whose logFC > 2. e Relative CLDN6 mRNA expression in eight different gastric cell lines. f Relative CLDN6 mRNA expression in gastric tissues (n = 16 in each group, student’s t test). g Relative CLDN6 protein expression in gastric tissues. h, i. Relative CLDN6 mRNA expression in GSE26942 and GSE54129 datasets (student’s t test). Each experiment was performed three times and measurement data were presented as the mean ± SD. ns, non-significant; *p < 0.05, **p < 0.01; ***p < 0.001.

Fig. 2. Increased CLDN6 expression is associated with worse prognosis in GC patients.

a Proportion of different TNM stage in low CLDN6 and high CLDN6 expression groups (Spearman’s rank correlation). b Overall survival analysis of TCGA dataset by Kaplan-Meier Plotter (TCGA database: low CLDN6 group n = 189, high CLDN6 group n = 189). c. CLDN6 protein expression in GC from tissue microarray (scale bar = 100 µm). d Different IHC score of different pathological stages in GC (Spearman’s rank correlation). e. Overall survival analysis of tissue microarray by Kaplan–Meier Plotter (GC tissue microarray: low CLDN6 group n = 239, high CLDN6 group n = 255). Each experiment was performed three times and measurement data were presented as the mean ± SD. ns, non-significant; *p < 0.05;**p < 0.01; ***p < 0.001.

Fig. 3. CLDN6 promotes proliferation and invasion abilities of GC in vitro.

a, b Colony formation assays showed the effects of CLDN6 knockdown in MKN28 and AGS gastric cancer cell growth (student’s t test, scale bar = 1 cm). c Cell viability of MKN28 and AGS was measured by CCK8 assay (student’s t test). d, e Wound healing assay showing the effect of silencing CLDN6 expression in MKN28 and AGS cells (student’s t test, scale bar = 100 µm). f, g Transwell assay showing the migration ability of MKN28 and AGS cells after knockdown of CLDN6 expression (student’s t test, scale bar = 50 µm). Each experiment was performed three times and measurement data was presented as the mean ± SD. ns, non-significant; *p < 0.05;**p < 0.01; ***p < 0.001.

Fig. 4. CLDN6 promotes proliferation and invasion abilities of GC in vivo.

a Nude mice were injected with sh-CLDN6 or sh-control vector transfected cells (sh-CLDN6 [n = 5] or sh-control [n = 5], scale bar = 1 cm). b Tumor weight of sh-control and sh-CLDN6 groups (student’s t test). c Subcutaneous tumors were stained with Ki67 in sh-CLDN6 and sh-control group (scale bar = 50 µm). d Tumor volume of sh-control and sh-CLDN6 groups (student’s t test). e Tumor growth assessment in liver metastasis model of GC (red arrows indicate metastasis, scale bar = 1 cm). f Tumor numbers in sh-control and sh-CLDN6 groups in liver metastasis model of GC (n = 5 in each group, student’s t test). g E-cadherin and N-cadherin staining of adjacent normal tissues (liver tissue group) and tumor tissues (sh-CLDN6 and sh-control groups, scale bar = 50 µm). Each experiment was performed three times and measurement data was presented as the mean ± SD. ns, non-significant; *p < 0.05;**p < 0.01; ***p < 0.001.

Fig. 5. CLDN6 promotes YAP1 nuclear translocation that interacts with snail1 to promote the process of EMT.

a GSEA analysis was performed for CLDN6 expression in GC, including low CLDN6 expression group (40 samples) and high CLDN6 expression group (40 samples) from GSE110875. b, c Immunoprecipitation assay showed the interaction between CLDN6 and LATS1/2. d Protein levels of CLDN6, LATS, p-LATS, YAP1, p-YAP1, zeb1, snail1, twist1, and β-actin were measured in MKN28 and AGS cells by western blotting. e Cellular immunofluorescence assay showed YAP1 localization in MKN28 GC cells in sh-control group and sh-CLDN6 group (scale bar = 20 µm). f, g Nuclear extraction experiment showed the quantification of the cytoplasm/nucleus of YAP1. h, i, j Immunoprecipitation assay showed the interaction between YAP1 and snail1 in MKN28 and AGS cells. k mRNA levels of downstream genes of YAP1 in sh-control and sh-CLDN6 groups by RT-PCR. l mRNA levels of CDH1, CDH2, and Vimentin in sh-control and sh-CLDN6 groups by RT-PCR. m Protein levels of E-cadherin, N-cadherin, and Vimentin in sh-control and sh-CLDN6 groups by western blotting. n Schematic of primer design for CDH1 promoter sequences. Ten primer sets with a 300-bp partition were designed for PCR to test the direct binding of YAP1-snail1 to the CDH1 promoter and the primer pairs produced 10 fragments of 300 bp. o Amplification of CDH1 promoter sequence from ChIP DNA validated the binding of YAP1-snail1 to the CDH1 promoter site. Every experiment was performed three times and measurement data was presented as the mean ± SD. ns, non-significant; *p < 0.05;**p < 0.01; ***p < 0.001.

Fig. 6. Overexpression of YAP1 partially rescued the diminished proliferative and invasive ability caused by silencing CLDN6 expression.

a Protein levels of CLDN6, LATS, p-LATS, YAP1, p-YAP1, snail1, and β-actin were measured in MKN28 and AGS cells by western blotting in sh-control, sh-CLDN6 and sh-CDLN6 + oe-YAP1S127A group. b Subcutaneous tumors in each group were stained with E-cadherin, N-cadherin and Ki67 (scale bar = 50 µm). c Nude mice were injected with sh-control vector, sh-CLDN6 and sh-CDLN6 + oe-YAP1S127A transfected cells (sh-control [n = 5], sh-CLDN6 group [n = 5], sh-CDLN6 + oe-YAP1S127A [n = 5], scale bar = 1 cm). d Tumor growth assessment in liver metastasis model of GC in different groups (n = 3 in each group, black arrows show metastasis, scale bar = 1 cm). e, f Tumor volume and weight of sh-control, sh-CLDN6, and sh-CDLN6 + oe-YAP1S127A groups (student’s t test). g Liver metastatic tissues were stained for E-cadherin or N-cadherin (sh-CLDN6 group had no liver metastasis, scale bar = 50 µm). h Colony formation assay showed the effects of snail1 overexpression on MKN28 and AGS gastric cancer cell growth (student’s t test, scale bar = 1 cm). i Transwell assay showed the migration ability of MKN28 and AGS cells after snail1 overexpression (student’s t test, scale bar = 50 µm). j Wound healing assay showed the influence of snail1 overexpression on MKN28 and AGS cells (student’s t test, scale bar = 100 µm). Every experiment was performed three times and measurement data was presented as the mean ± SD. ns, non-significant; *p < 0.05; **p < 0.01; ***p < 0.001.

Fig. 7. Schematic representation of the mechanism of CLDN6-YAP1-snail1 axis.

Upregulated CLDN6 expression: CLDN6 interacted with LATS, reducing the conversion of LATS into p-LATS, and YAP1 into p-YAP1. Therefore, increased amount of YAP1 entered cell nucleus to activate its downstream target genes. Increased YAP1 interacted with snail1 to bind to the promoter of EMT related genes to enhance EMT progression. Downregulated CLDN6 expression: Decreased amount of CLDN6 interacted with LATS thereby increasing p-LATS and resulting in increased p-YAP1. Due to decreased levels of YAP1 in the cell nucleus, the downstream pathways were inhibited.