KCTD11 inhibits progression of lung cancer by binding to β‐catenin to regulate the activity of the Wnt and Hippo pathways

Abstract KCTD11 has been reported to be a potential tumour suppressor in several tumour types. However, the expression of KCTD11 and its role has not been reported in human non‐small cell lung cancer (NSCLC). Whether its potential molecular mechanism is related to its BTB domain is also unknown. The expression of KCTD11 in 139 NSCLC tissue samples was detected by immunohistochemistry, and its correlation with clinicopathological factors was analysed. The effect of KCTD11 on the biological behaviour of lung cancer cells was verified in vitro and in vivo. Its effect on the epithelial‐mesenchymal transition(EMT)process and the Wnt/β‐catenin and Hippo/YAP pathways were observed by Western blot, dual‐luciferase assay, RT‐qPCR, immunofluorescence and immunoprecipitation. KCTD11 is under‐expressed in lung cancer tissues and cells and was negatively correlated with the degree of differentiation, tumour‐node‐metastasis (TNM) stage and lymph node metastasis. Low KCTD11 expression was associated with poor prognosis. KCTD11 overexpression inhibited the proliferation and migration of lung cancer cells. Further studies indicated that KCTD11 inhibited the Wnt pathway, activated the Hippo pathway and inhibited EMT processes by inhibiting the nuclear translocation of β‐catenin and YAP. KCTD11 lost its stimulatory effect on the Hippo pathway after knock down of β‐catenin. These findings confirm that KCTD11 inhibits β‐catenin and YAP nuclear translocation as well as the malignant phenotype of lung cancer cells by interacting with β‐catenin. This provides an important experimental basis for the interaction between KCTD11, β‐catenin and YAP, further revealing the link between the Wnt and Hippo pathways.

interact with transcription factors (such as TEADs) to enhance the transcriptional activity of downstream effector molecules and promote cell proliferation. [5][6][7] Wnt signalling pathway is also involved in many biological processes, including tissue and organ growth, cell proliferation, apoptosis and stem cell maintenance. [8][9][10] β-catenin acts as a transcriptional activator of the canonical Wnt signalling pathway. When the canonical Wnt pathway is stimulated by Wnt signalling, β-catenin cannot be degraded by a complex composed of GSK-3β, Axin and APC; thus, its nuclear translocation is increased, allowing it to interact with the transcription factor TCF/LEF to promote the transcriptional activity of downstream target genes. 11,12 Recent studies have shown that there is a complex crosstalk between the Hippo / YAP and Wnt/β-catenin signalling pathways.
Specifically, YAP/TAZ combined with β-catenin to block β-catenin nuclear localization is the core mechanism of how the Hippo pathway inhibits the Wnt pathway. 13,14 Park et al. found that YAP/TAZ, as a downstream effector of the Wnt signalling pathway, can be induced and activated by Wnt5a/b and Wnt3a, which enhance TEAD-mediated transcription activities. 15 Azzolin et al. found that APC directly regulated the degradation of YAP/TAZ through the β-catenin degradation complex. 16 Studies by Konsavage et al. showed that knockdown of βcatenin reduces YAP mRNA and protein levels. 8 Elucidating the mechanisms of mutual regulation of these two pathways may reveal new directions for the development of tumour-targeted drugs.
KCTD11 is a member of the potassium channel tetramerization domain (KCTD) family. Studies have shown that the highly conserved BTB domain (Bric-a-brac/Tramtrack/Broad complex) is involved in complex intracellular signalling. 17,18 Cullin 3 (Cul3), a scaffold protein involved in the degradation of a variety of intracellular proteins, is a major regulator of the cell cycle and developmental processes. 19,20 The N-terminal BTB domains of some KCTD proteins can be used as bridges connecting Cul3 and substrates and play the role of Cul3 ubiquitin ligase. They have multiple biological functions and are closely related to protein ubiquitination. 21,22 However, whether KCTD11 can promote the degradation of Hippo/YAP and Wnt/β-catenin and the mechanism by which it may exert tumour inhibition is unknown. In this study, we first verified the level of KCTD11 expression in lung cancer tissues and cells. We also investigated the effects of KCTD11 on the proliferation and invasion of lung cancer cells. We demonstrated that KCTD11 inhibits β-catenin expression and directly binds to β-catenin. Therefore, we aimed to investigate whether KCTD11 can regulate the Wnt pathway by binding to β-catenin via the BTB domain, thereby regulating the Hippo pathway, as well as clarify the role of this complex in crosstalk between the Hippo and Wnt signalling pathways.

| Patients and specimens
After obtaining informed consent from the local medical examination committee of China Medical University, tissue samples from 139 patients with a diagnosis of non-small cell lung cancer (78 males and 61 females) were obtained from those who underwent surgical resection at the First Affiliated Hospital of China Medical University from 2014 to 2017. The average age of the patients was 60 years. None of the patients underwent either radiotherapy or chemotherapy before surgical resection and received standard chemotherapy after surgery.
According to the 2015 World Health Organization classification guidelines for lung tumours, 23 immunohistochemical staining was used to assess histological type and degree of differentiation. The sample comprised 68 cases of squamous cell carcinoma and 71 cases of adenocarcinoma, of which 79 cases were classified as highly differentiated and 60 cases were classified as moderately or poorly differentiated.
According to the pathological tumour lymph node metastasis (TNM) staging of the International Union against Cancer (seventh edition) (Detterbeck et al., 2017), specimens can be classified into stages I-II (n = 73) and III (n = 66). Lymph node metastases occurred in 62 of 139 patients. In addition, a total of 20 newly isolated specimens (including tumours and normal tissues) were collected from surgical resection and immediately stored at 80℃ to extract tissue proteins.

| Immunohistochemistry (IHC)
The analysis was performed as previously described. 24 Briefly, tissue sections were cultured with KCTD11 rabbit polyclonal antibody (1:100 dilution; Sigma-Aldrich). Two independently blinded investigators examined all tumour sections by taking five random fields from each section, with 100 cells in the field of view observed and magnified 400 times for scoring. Due to differences in the lesions, the proportion of positive cells and staining intensity were considered. The KCTD11 staining positive cell rate was assigned as follows: 1 (1%-25%), 2 (26%-50%), 3 (51%-75%) and 4 (76%-100%). The staining intensity was divided into 0 (no staining), 1 (weak staining, light yellow staining), 2 (medium staining, yellow staining) or 3 (strong staining, brown staining). The two scores for each tumour sample were multiplied to give a final score of 0-12, with a tumour sample score ≥4 defined as positive expression, scores of 1-4 defined as low expression and a score of 0 defined as negative expression. Phosphate buffer (MaiXin) and goat serum (MaiXin) were used as negative controls.

| Colony formation, Transwell and MTT assays
After 24 h of transfection with plasmids or siRNA, the cells were seeded in a 6 cm cell culture dish (1000 cells per dish) and incubated for 10 days. When the number of single colony cells reached approximately 50, cells were washed with PBS, fixed with 4% paraformaldehyde and stained with crystal violet, and the number of colonies was statistically analysed.

| Transwell migration assay
A cell migration assay was performed using a 24-well Transwell chamber with an 8-pore size (Costar). After 24 h of transfection, the cells were counted by trypsinization, 100 μl of serum-free medium containing 3 × 10 5 cells was evenly spread to the upper chamber while a medium containing 10% FBS as a chemical attractant was added to the lower chamber, after which the samples were cultured for 16 h. In the 4% paraformaldehyde-fixed cells, after clearing the non-migrating cells in the upper chamber using a cotton swab, crystal violet staining was performed. The number of migrated cells was counted under a microscope, and ten high-power fields were randomly selected.

| MTT assays
Cell counts were measured 24 h after transfection, and cells were plated in 96-well plates in media containing 10% FBS at approximately 3,000 cells/well. Cell viability was determined after five consecutive days. Briefly, 20 μl of 5 mg/ml MTS solution was added to each well in the dark and cultured for 2 h, and the results were spectrophotometrically determined using a test wavelength of 490 nm.

| Dual-luciferase assay
Luciferase activity in cell extracts was determined using a dualluciferase reporter assay kit (Promega). The reporter activity was normalized to co-express β-galactosidase activity. All luciferase plasmids were purchased from Addgene. The transcriptional activity of YAP/TEAD in the Hippo pathway was determined using the

| RNA extraction and real-time RT-PCR (RT-qPCR)
Twenty-four hours after cell transfection, RNA was extracted and subjected to RT-qPCR analysis as described previously (Imajo et al., 5′-GGCTGTTGTCATACTTCTCATGG-3′ (reverse). Six weeks after inoculation, the mice were sacrificed and an autopsy was performed to examine the growth and spread of the tumour.

| Transplantation of tumour cells into nude mice
The excised tumour tissue was fixed in 4% formaldehyde (Sigma) and embedded in paraffin. After H&E staining, tumour clusters in the lung tissue were analysed under a microscope.

| Statistical analysis
Statistical analysis was performed using the statistical software SPSS 22.0, which was used to assess the correlation between KCTD11 expression and clinicopathological factors. Prognostic value was tested using a Cox regression model. All clinical pathology parameters were included in the Cox regression model and were tested by univariate analysis using the enter method and by multivariate analysis using a forward stepwise logistic regression method. Image J was used for image analysis of the Western blot results. Differences between test groups were compared using a paired t test using GraphPad Prism software. Differences were considered statistically significant at p < 0.05.

| KCTD11 was downregulated in NSCLC tissues and cells
We examined the expression of KCTD11 in NSCLC tissues and cells. Western blotting showed that the expression of KCTD11 was significantly higher in 20 cases of normal tissues than in paired NSCLC tissues ( Figure 1A, B). Immunohistochemical staining also showed that KCTD11 was negatively expressed in 54.7% of NSCLC tissues (76/139, Table 1). KCTD11 was weakly expressed in the nuclei and cytoplasm or negatively expressed in NSCLC tissues, in contrast to its strong expression in normal tissues ( Figure 1C). The protein expression of KCTD11 was also higher in HBE cells than in NSCLC cell lines (H1299, LK2, A549, H661, H1299 and H292) ( Figure 1D).

| Expression of KCTD11 correlated with clinical factors
We investigated the relationship between KCTD11 expression and clinicopathological factors in patients with NSCLC. The low KCTD11 expression was significantly correlated with the degree of differentiation (p = 0.033), advanced pathological TNM (pTNM) stage (p = 0.044) and positive lymph node metastasis (p = 0.015), but not with age, sex and histological type (Table 1). Moreover, Kaplan-Meier survival analysis showed that the overall survival rate of the KCTD11-positive group was significantly higher than that of the KCTD11-negative group (p < 0.001, log-rank test), suggesting an association between KCTD11 expression and prognosis ( Figure 1E). Univariate analysis showed that KCTD11, high TNM classification and positive lymph node metastasis were significant prognostic factors for NSCLC (low KCTD11 expression: hazard ratio    Figure S1A and D and E; Figure S1B and E). The MTT assay showed that the proliferation rate of the A549 and H460 cell lines after overexpression of KCTD11 was significantly lower than that of the control group (p < 0.05). Consistently, the proliferation rate was significantly reduced in KCTD11-depleted H1299 cells and HBE cells (p < 0.05).

| KCTD11 inhibited proliferation and invasion of HBE and NSCLC cells in vitro and in vivo
( Figure 2C and F; Figure S1C and F).
We  Figure 2J and Figure S1H). The KCTD11 protein levels are shown in Figure 2K and L. Collectively, these results indicate that KCTD11 regulates the proliferation and invasion of NSCLC cells in vivo.

| KCTD11 activated the Hippo pathway by upregulating the phosphorylation level of YAP
The dual-luciferase assay showed that transfection of KCTD11 upregulated the activity of the Hippo signalling pathway, and that depletion of KCTD11 inhibited the activity of the Hippo signalling pathway ( Figure 3A, B). We then found that CTGF, CYR61 and cyclin E mRNA, which are related to the Hippo signalling pathway, were downregulated in KCTD11-overexpressing A549 cells and upregulated in KCTD11-depleted H1299 cells ( Figure 3C, D), suggesting that KCTD11 is involved in the Hippo signalling pathway.  Figure 3E, F).
We also found that nuclear translocation of YAP was inhibited in A549 cell lines after transfection with KCTD11, which can be demonstrated by nucleus-cytoplasm isolation and immunofluorescence ( Figure 3G and I). These results were reversed in the H1299 cell line, with depletion of KCTD11 promoting the nuclear translocation of YAP ( Figure 3H and J). Therefore, these results indicate that KCTD11 activates the Hippo signalling pathway by promoting the phosphorylation of YAP.

| KCTD11 inhibits Wnt pathway and nuclear translocation of β-catenin
According to previous reports, crosstalk exists between the Wnt and Hippo pathways. Therefore, to further explore the molecular mech-

| KCTD11 binds to β-catenin via the BTB domain
Recent studies have shown that the BTB domain of the KCTD family is a highly versatile scaffold that acts as a bridge to substrates involved in the ubiquitination of multiple proteins (Wang et al., 2016). The relationship between KCTD1 and β-catenin has been demonstrated previously (Li et al., 2014). We speculated that the BTB domain of KCTD11 might interact with β-catenin. We performed immunohistochemical staining of serial sections and statistical analysis to show that KCTD11 expression was correlated with β-catenin membranous expression (p = 0.035) ( Table 3, Figure 6A). Immunofluorescence staining also revealed that the two proteins were co-localized in the NSCLC cell lines ( Figure 6B).
Co-immunoprecipitation was performed to detect the interaction between endogenous KCTD11 and β-catenin in H1299 cells ( Figure 6C). The interaction between exogenous KCTD11 and β-catenin was also detected by transfection with Myc-tagged KCTD11 plasmids in A549 cells ( Figure 6D). We constructed a KCTD11 mutant plasmid (MYC-KCTD11-ΔBTB) and found that MYC-KCTD11-ΔBTB could not bind to β-catenin in A549 cells ( Figure 6E, F). Based on the above results, we concluded that KCTD11 interacts with β-catenin via its own BTB domain.

| KCTD11 regulates Hippo pathway activity by β-catenin
Our experiments showed that KCTD11 negatively regulates the transmission of the Wnt signalling pathway and suppresses the expression of β-catenin. Further investigation revealed that the levels of active β-catenin were downregulated, and that Pβ-catenin was upregulated in KCTD11-overexpressing A549 and H460 cells, whereas the results were reversed in KCTD11-depleted H1299 cells and HBE cells ( Figure 7A, B). It has been reported that YAP is a direct target gene of Wnt /β-catenin, and that the deletion of β-catenin inhibits the expression of YAP [14]. Interestingly, we co-transfected KCTD11 plasmid and siβ-catenin in the A549 cell line, but P-YAP was not upregulated ( Figure 7C), indicating that the stimulatory effect on YAP of KCTD11 was dependent on β-catenin.
To further investigate whether the tumour-suppressive effect of KCTD11 is mediated by the BTB domain, we transfected a WT KCTD11 plasmid and its mutant MYC-KCTD11-ΔBTB in A549 cells.
Western blot analysis and dual-luciferase assay confirmed that the BTB domain is a KCTD11 functional region, which affects the role of KCTD11 in tumour suppression ( Figure 7D-F).

| DISCUSS ION
Lung cancer, which is one of the leading causes of cancer-related deaths worldwide, is still on the rise. 25 Non-small cell lung cancer accounts for approximately 80% of all lung cancer cases. Although treatment techniques (surgical resection, chemotherapy and radiotherapy) are improving, overall long-term survival is poor. 26  By co-immunoprecipitation and immunofluorescence, we found that KCTD11 can bind to β-catenin and co-localize in lung cancer tissues and cells. When KCTD11 was transfected and interfered with βcatenin concurrently, we found that YAP and P-YAP protein levels decreased with reduced β-catenin protein levels compared with single KCTD11 transfection. Overall, the present study found that KCTD11 binds to β-catenin and inhibits β-catenin nuclear translocation which further inhibits the Hippo pathway, thus leading to decreased proliferation and metastasis of lung cancer cells. In further studies, we directly knocked out the BTB domain and found that KCTD11 cannot bind to β-catenin, thus losing its role as a tumour suppressor.

| CON CLUS ION
The results of this study indicate that the human potassium channel tetramer protein KCTD11 can bind to β-catenin through its BTB domain in NSCLC cells, inhibit the Wnt pathway, increase Hippo pathway activity, and inhibit the proliferation and migration of lung cancer cells. Therefore, KCTD11 constitutes the link between the Wnt and Hippo pathways, and may be a potential target for lung cancer drug development.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no competing interests.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.

O RCI D
Xu-Yong Ln https://orcid.org/0000-0001-6714-0639 F I G U R E 7 KCTD11 regulates Hippo pathway activity by β-catenin. (A) Transfection of KCTD11 in A549 and H460 cell lines, Western blot showing that the expression level of active β-catenin is downregulated and Pβ-catenin is upregulated. (B) The expression level of active β-catenin is upregulated, and Pβ-catenin is downregulated after silencing of KCTD11 in H1299 and HBE cells. (C) KCTD11 inhibits YAP through β-catenin. A549 cells are transfected with KCTD11 and EV, and si-catenin and siNC. Western blot showing that after knocking out β-catenin, the expression levels of YAP and P-YAP are reduced and that overexpression of KCTD11 could not reverse this change, with the level of YAP and P-YAP still being inhibited. (D, E) KCTD11 acts as a tumour suppressor through the BTB domain. A549 cells are transfected with EV, KCTD11 and mut-ΔBTB plasmid and the level of proteins are detected by Western blot. Overexpression of mut-ΔBTB does not affect the expression of Hippo and wnt pathway-related proteins and EMT-related proteins. (F) The Wnt TOPflash reporter and Hippo pGTII luciferase reporter are used to detect the transcriptional activity of YAP / TEAD and β-catenin / TCF-4 after transfection of mut-ΔBTB. Abbreviations: EV, empty vector; siNC, negative control siRNAs