Suppression of long non‐coding RNA TNRC6C‐AS1 protects against thyroid carcinoma through DNA demethylation of STK4 via the Hippo signalling pathway

Abstract Objectives Thyroid carcinoma (TC) represents a malignant neoplasm affecting the thyroid. Current treatment strategies include the removal of part of the thyroid; however, this approach is associated with a significant risk of developing hypothyroidism. In order to adequately understand the expression profiles of TNRC6C‐AS1 and STK4 and their potential functions in TC, an investigation into their involvement with Hippo signalling pathway and the mechanism by which they influence TC apoptosis and autophagy were conducted. Methods A microarray analysis was performed to screen differentially expressed lncRNAs associated with TC. TC cells were employed to evaluate the role of TNRC6C‐AS1 by over‐expression or silencing means. The interaction of TNRC6C‐AS1 with methylation of STK4 promoter was evaluated to elucidate its ability to elicit autophagy, proliferation and apoptosis. Results TNRC6C‐AS1 was up‐regulated while STK4 was down‐regulated, where methylation level was elevated. STK4 was verified as a target gene of TNRC6C‐AS1, which was enriched by methyltransferase. Methyltransferase’s binding to STK4 increased expression of its promoter. Over‐expressed TNRC6C‐AS1 inhibited STK4 by promoting STK4 methylation and reducing the total protein levels of MST1 and LATS1/2. The phosphorylation of YAP1 phosphorylation was decreased, which resulted in the promotion of SW579 cell proliferation and tumorigenicity. Conclusion Based on our observations, we subsequently confirmed the anti‐proliferative, pro‐apoptotic and pro‐autophagy capabilities of TNRC6C‐AS1 through STK4 methylation via the Hippo signalling pathway in TC.


| INTRODUC TI ON
As the most common malignancy of the endocrine organs, thyroid carcinoma (TC) often manifests itself as a tumour afflicting the thyroid epithelium. Studies have highlighted a rapid increase in TC incidence worldwide (>5% annually) with reports indicating that women tend to be much more susceptible than their male counterparts. 1,2 There are two differentiated subtypes of TC malignancies namely, follicular thyroid carcinoma and papillary thyroid carcinoma, both of which generally require extended periods to heal or to progress. 3 Generally speaking, patients suffer from TC often undergo surgical treatment, including total thyroidectomy, and radioactive iodine both of which are widely considered to be a standard treatment option in clinical practice. That being said, both total thyroidectomy as well as radioactive iodine procedures can result in regional recurrence, which may lead to a reduction in life quality. 4 Thus, new progressive therapies are required in order to more effectively target thyroid tumours. LncRNAs represent RNAs longer than 200 nucleotides, which have been linked to various biological processes, including that of nuclear import and RNA transcription.
Our ever-expanding knowledge of long non-coding RNAs (lncRNAs) represents an opportunity to identify new therapeutic targets.
The altered expression of lncRNAs has been shown to be a critical process in thyroid carcinogenesis and has been linked to with patients having more adverse prognoses. 5 The functions of lncRNAs have been documented in relation to their involvement in gene regulation at a post-transcriptional level. 6 For example, the lncRNA NAMA has been revealed as a downstream target gene of the mitogen-activated protein kinase pathway with reports linking it to cellular-growth arrest in papillary thyroid carcinoma. 7 Serine/ threonine-protein kinase 4 (STK4), also referred to as mammalian sterile 20-like kinase-1 and Krs-2, is a 56-60-kDa protein that has been shown to be down-regulated in various cancers, including that of prostate cancer and hepatocellular carcinoma. 8,9 In addition, existing literature has suggested that DNA hypermethylation and post-translational modification can mediate the loss of STK4 activity. 10 Ready et al concluded that the dysregulation of STK4 is linked to carcinogenesis associated with poor outcomes in addition to implicating the Hippo signalling pathway in this process. 11 The Hippo signalling pathway, comprised of mammalian STE20-like protein kinase 1 and 2 (MST1/2), large tumour suppressor kinase 1/2 (LAST1/2) and its downstream transcription co-activator, Yesassociated protein (YAP), have been implicated in the regulation of tumour suppression concerning tissue repair and regeneration. 12 Based on the exploration of the aforementioned literature, we asserted the hypothesis that lncRNA TNRC6C-AS1 could potentially work in tandem with STK4 and the Hippo signalling pathway in TC cells, whereby tumour progression can be regulated.

| Ethics statement
The study was conducted with the approval of the Medical Ethics Committee of Suqian First Hospital and Huai'an Hospital Affiliated to Xuzhou Medical College and Huai'an Second People's Hospital.
All animal experiments were performed in strict accordance with the approval of the Animal Protection and Use Committee. All participants signed written informed consents.

| Screening of differentially expressed genes in TC
Differentially expressed genes (DEGs) in TC were downloaded from the TCGA (http://cancergenome.nih.gov/) database, with R software employed for statistical analysis. Differential analysis was conducted for the transcriptome profiling data with package edgeR of R. 13 Falsepositive discovery (FDR) correction was performed on P-values with the multitest package. FDR value of <0.05 and |log2 (fold change)| value of >2 were used as the threshold when screening out DEGs.

| Immunohistochemistry
The for 10 minute. PCR products were resolved using a 20 g/L agarose.
The results obtained were then observed using an automatic analysis system of electrophoresis gel imaging. Due to the fact that the methylated primers could amplify products while the unmethylated primers could not amplify products, the primers were identified as fully methylated, fully unmethylated and partly unmethylated. 17 The aforementioned method was applicable in an identical fashion to the detection among the cells. Each experiment was repeated three times.   final concentration: 100 U/mL). Following incubation at 37°C with 5% CO 2 , RT-qPCR was performed in an attempt to select the cell line with the highest expression levels of lncRNA TNRC6C-AS1. Cells at the logarithmic growth stage were selected for subsequent experimentation.

| Cell grouping and transfection
Thyroid carcinoma SW579 cells displaying logarithmic growth were seeded into a 6-well plate. When cell confluence reached 80%-90%, order to confirm successful transfection. 19 A transfection efficiency of >75% was considered to be reflective of successful transfection.

| Fluorescence in situ hybridization
The localization of TNRC6C-AS1 expression in SW579 cells was predicted using the website http://lncatlas.crg.eu/. TNRC6C-AS1 sub-cellular localization was detected using a fluorescence in situ hybridization (FISH) kit (Roche, Basel, Switzerland). Fluorescence images were obtained under a confocal laser scanning microscope (FV1000; Olympus, Tokyo, Japan). Each experiment was repeated three times.

| Luciferase reporter gene assay
The luciferase activity = firefly luciferase activity/renilla luciferase activity. Each experiment was conducted three times.

| RNA-binding protein immunoprecipitation
The experiment was performed in accordance with the instruc- RT-qPCR using TNRC6C-AS1-specific primers as depicted in Table 1.

| Chromatin immunoprecipitation
The enrichment analysis of DNMT1, DNMT2 and DNMT3 in STK4 promoter region was conducted using a chromatin immunoprecipitation (CHIP) kit (Millipore). The binding of DNMT1, DNMT2 and DNMT3 to STK4 promoter region was detected using specific primers in STK4 promoter region (Table 1). Each experiment was repeated three times.

| Immunofluorescence staining
Following conventional detachment and transfection, the cells were counted and cultured in immunofluorescence chambers at a density of 2 × 10 5 cells/well. When confluence had reached approximately 90%, the cells were washed three times with PBS on ice and fixed in 4% paraformaldehyde (1 mL/well) at room temperature for 15 minute, followed by three additional PBS washes. The cells were then treated with 0.3%

| 5-Ethynyl-2'-deoxyuridine assay
Thyroid carcinoma cells at the logarithmic growth phase were seeded into a 96-well plate with 2 × 10 3 to 4 × 10 4 cells in each well, which were then permitted to attach over a 24 hour period. Transfection was performed with triplicate wells in each group. Next, 100 μL fixative was added into each well and incubated for 30 minute at room temperature. Next, 2 mg/mL glycine was added into each well for another 5 minute of incubation. The plate was then washed with PBS (100 μL/well) for 5 minute. A penetrant (100 μL/well, PBS containing 0.5% Triton X-100) was subsequently added and incubated for

| Flow cytometry
The apoptosis of the SW579 cells following a 24 hour period of treat-

| Hoechst staining
A coverslip was placed on the 6-well plate. Following a 24 hour of treatment, HCT116 and SW579 cells were seeded into the 6-well plate at the density of 5 × 10 5 cells/well and cultured in an incubator overnight at 37°C with 5% CO 2 . When cell confluence had reached approximately 50%-80%, the cells were fixed with 0.5 mL fixative for 10 minute on the following day. After the fixative had been removed, the cells were

| Observation under a transmission electron microscope
After the cells had been collected, they were then fixed in a 2.

| Statistical analysis
All data were processed by SPSS 21.0 statistical software (IBM Corp. Armonk, NY, USA). Measurement data were expressed as mean ± standard deviation. The t test was used to compare data between two groups. One-way analysis of variance (ANOVA) was employed for comparisons among multiple groups. A P value of <0.05 was considered to be indicative of statistical significance.

| LncRNA TNRC6C-AS1 was highly expressed while STK4 was reduced in TC tissues
Initially, based on the analysis results of The Cancer Genome Atlas (TCGA), TNRC6C-AS1 was determined to be highly expressed in TC tissues while STK4 was poorly expressed. As depicted in Figure 1A, A negative correlation was found between TNRC6C-AS1 and STK4. RT-qPCR and Western blot analysis methods were performed in order to determine the expression levels of lncRNA TNRC6C-AS1 and STK4 in TC and adjacent normal tissues. The results revealed notably elevated lncRNA TNRC6C-AS1 expressions while the levels of STK4 were decreased in TC tissues when compared the adjacent normal tissues (P < 0.05; Figure 1B-D). Next, an immunohistochemistry assay was performed in order to detect the STK4-positive expression, the results of which indicated that the positive expression of STK4 was primarily in the cytoplasm, while the positive rate of STK4 was much lower in the TC tissues when compared to the adjacent normal tissues (P < 0.05; Figure 1E,F).
RT-qPCR methods were further conducted in an attempt to select the TC cell line with the highest TNRC6C-AS1 expression.
Compared with normal thyroid cell line Nthy-ori 3-1, the relative expression of TNRC6C-AS1 elevated in TC cell lines FRO, ARO, PDTC-1 and SW579, among which the cell line SW579 exhibited the highest TNRC6C-AS1 expression (P < 0.05; Figure 1G). On the basis of the aforementioned findings, the TC cell line SW579 was selected for subsequent experiments.

| STK4 was a downstream target gene of TNRC6C-AS1
The sub-cellular localization of TNRC6C-AS1 was predicted using the site http://lncatlas.crg.eu/, which revealed that TNRC6C-AS1 was located in the nucleus of multiple cell lines (Figure 2A). This finding was further verified by experimental FISH means, which demonstrated that TNRC6C-AS1 was primarily expressed in the nucleus ( Figure 2B). promoter region ( Figure 2C), suggesting that STK4 was indeed a target gene of TNRC6C-AS1. The targeting relationship between STK4 and TNRC6C-AS1 was further verified through the application of the luciferase reporter gene assay ( Figure 2D). The results indicated that the luciferase activity of STK4 wild type was significantly inhibited in the TNRC6C-AS1 group when compared to that in the NC group (P < 0.05), while the luciferase activity of mutant 3′-UTR did not exhibit any significant change (P > 0.05). These results indicated that TNRC6C-AS1 could directly target STK4.
In order to ascertain as to whether TNRC6C-AS1 could influence STK4 expression, RT-qPCR and Western blot analysis methods were conducted following transfection, the results of which are shown in Compared with the si-TNRC6C-NC group, the opposite tendency was observed in the si-TNRC6C group (P < 0.05). All the obtained data indicated that STK4 expression could be regulated by TNRC6C-AS1.

| TNRC6C-AS1 inhibited STK4 expression through DNA methylation
MethPrimer software was applied for the analysis of CpG island in STK4 promoter region by inputting nucleotide sequence around 3000 bp close to the STK4 promoter region. The results revealed that STK4 promoter region was located in the CpG island ( Figure 3A), indicating that STK4 expression was indeed affected by promoter methylation. In order to verify as to whether TNRC6C-AS1 was associated with the methylation level of STK4 promoter region, 78 cases of TC samples were randomly selected with the methylation level of the CpG island in the STK4 promoter region of the selected samples were detected using methylation-specific polymerase (MSP). The results revealed that the methylation level of STK4 was notably elevated in the TC tissues compared to that of the adjacent normal tissues (P < 0.05; Figure 3B). The obtained results suggested that the STK4 expression decreased as the methylation level increased, which provided evidence proving that STK4 was poorly expressed in TC tissues ( Table 2). RIP was then employed to detect the relative TNRC6C-AS1 enrichment binding to IgG, DNMT1, DNMT2 and DNMT3, respectively. After being normalized to IgG, the relative TNRC6C-AS1 enrichment binding to DNMT1, DNMT2 and DNMT3 was notably increased, which demonstrated that TNRC6C-AS1 promoted DNMT1, DNMT2 and DNMT3 enrichment ( Figure 3C). CHIP was then performed in order to detect the relative methyltransferase enrichment in STK4 promoter region in adjacent normal and TC tissues. The results revealed the existence of methyltransferase enrichment in the STK4 promoter region in TC tissues, while limited enrichment was observed in the adjacent normal tissues (P < 0.05) ( Figure 3D).

MSP was employed in order to detect the methylation level of
CpG island on STK4 promoter in TC cells with either highly expressed or poorly expressed TNRC6C-AS1. The results revealed that the CpG island in STK4 possessed a higher methylation level in the cells with over-expressed TNRC6C-AS1 while reductions in STK4 expression were noted, vice versa. DNA methyltransferase inhibitor 5-Aza-CdR was introduced to promote the demethylation of STK4 promoter region in order to investigate the regulatory mechanism between TNRC6C-AS1 and SKT4. The obtained MSP results revealed that STK4 methylation was higher in the OE-TNRC6C + DMSO group but lower in the OE-TNRC6C + 5-Aza-CdR group (P < 0.05; Figure 3E). Through CHIP, an evident reduction was found in relation to the relative methyltransferase enrichment in STK4 promoter region of the cells in the OE-TNRC6C + 5-Aza-CdR group compared with the OE-TNRC6C + DMSO group, inferring that 5-Aza-CdR could reverse the STK4 methylation caused by TNRC6C-AS1 over-expression in TC cells ( Figure 3F). The above-mentioned findings provided elucidated indicating that the methylation of CpG island in the STK4 promoter region shares a significant correlation to TNRC6C-AS1 expression.

| STK4 modulated the Hippo signalling pathway
The effects of STK4 on the Hippo signalling pathway were investigated through the application of a Western blot analysis ( Figure 4A,B).
The results revealed that compared with the NC group, the OE-STK4 group exhibited elevated levels of MST1, LAST1 and LAST2, as well as increased YAP1 phosphorylation (P < 0.05). An opposite trend was observed in the si-STK4 group (P < 0.05). TA B L E 2 Methylation level of STK4 in thyroid carcinoma and adjacent normal tissues Immunofluorescence staining was performed in order to detect the nuclear transfer of YAP1 in SW579 cells. As illustrated in

| Up-regulation of STK4 inhibited proliferation while promoting apoptosis and autophagy via the Hippo signalling pathway
RT-qPCR and Western blot analysis were conducted to determine the expression of factors related to TC cell proliferation, apoptosis and autophagy. As depicted in Figure

| Silencing of TNRC6C-AS1 inhibited tumorigenesis in nude mice
The   21 Accumulating evidence has demonstrated the potential role of lncRNAs in TC progression and development. For example, the lncRNA PVT1 has been reported to contribute to the tumorigenesis of TC. 22 In the present study, we investigated the role of the particular long non-coding RNA TNRC6C-AS1 in TC. Our results demonstrated that the silencing of lncRNA TNRC6C-AS1 could suppress TC cell proliferation and tumorigenesis while promoting apoptosis and autophagy through the inhibition of STK4 methylation via the Hippo signalling pathway.
A key initial observation of our study revealed that STK4 was a downstream target gene of lncRNA TNRC6C-AS1. However, lncRNA TNRC6C-AS1 could be enriched by DNMT1, DNMT2 and DNMT3 and might regulate STK4 methylation level through binding to CpG island of STK4 promoter. DNA methylation has been defined as an epigenetic mechanism of gene regulation with reports suggesting it may be associated with mammalian development and disease. 23 A previously conducted functional study asserted that the methylation in CpG island of promoters results in silencing of its downstream genes at a transcriptional level, or the inactivation of tumour suppressor genes in specific cancers. 24 As a maintenance methyltransferase, DNMT1 is of great significance from a cell survival and tumorigenesis perspective, particularly in the cytosine hypermethylation in CpG islands of tumour suppressor genes. 25 In addition, reports have suggested that when STK3/4 is inactivated, YAP can translocate to the nucleus and increase the gene expression of its target, which off course was consistent with our results whereby YAP1 translocation from the cytoplasm to the nucleus was enhanced in the si-STK4 group. 26 The Hippo signalling pathway has been speculated to result in the phosphorylation of TAZ (transcriptional co-activator with PDZ-binding motif), a co-activator of YAP/transcription. Thus, their respective translocation to the nucleus could be prevented when gene activation is avoided and cell apoptosis is inhibited. 27 Our study also demonstrated that silencing of lncRNA TNRC6C-AS1 or the up-regulation of STK4 could inhibit proliferation and F I G U R E 6 Silencing of TNRC6C-AS1 inhibited tumorigenesis in nude mice. Panel A, images of resected tumours in each group. Panel B, tumour volumes on 7, 14, 21 and 28 d in each group. Panel C, tumour weights in each group. N = 8; *P < 0.05 vs. the OE-TNRC6C-NC, si-TNRC6C-NC and OE-STK4 + DMSO groups; the data are analysed using one-way ANOVA; the experiment was repeated three times Researchers have demonstrated that STK4 knockout mice exhibit enhanced levels of apoptosis, although STK4 deficiency in humans may elevate the degree of susceptibility to recurrent infections. 26 More importantly, the loss of STK3/STK4 has been documented to lead to autophagic structure accumulation and increased LC3II expression, which may further affect cells' ability to fight against infection. 28 Evidence has been presented suggesting that the LC3 family plays a role in the biogenesis and completion of autophagosomes. 26 The Hippo signalling pathway plays a crucial role in tissue growth through its downstream effector, YAP1. 29 Inactivated YAP1 has been speculated to function as a therapeutic target, owing to its ability to reduce tumour volume while inhibiting invasion and metastasis in mice with TC. 30 Likewise, similar experimental data obtained in a prior study published by Liu et al 27 revealed that the expression levels of autophagy markers, including Beclin-1, LC3I and LC3II, were reduced in thyroid papillary carcinoma cells.

| CON CLUS ION
In conclusion, the key findings presented in our study suggest that the over-expression of lncRNA TNRC6C-AS1 is associated with the development of TC through the modulation of cell apoptosis and autophagy by promoting STK4 methylation via inhibition of the Hippo signalling pathway (Figure 7). Specific methylation in cancer cells has been highlighted as a biomarker owing to its high sensitivity to aberrant methylation in cancer. 24 This being said, our study potentially provides a theoretical basis for which lncRNA TNRC6C-AS1 may be premised upon as a therapeutic target for TC treatment. However, due to the limited sample size and experimental conditions of this study, further studies are required in order to further our understanding in relation to the role of lncRNA TNRC6C-AS1 in TC.

CO N FLI C T O F I NTE R E S T
The authors hereby disclose no conflicts of interest.

O RCI D
Shao-Gang Ma https://orcid.org/0000-0002-6355-6415 F I G U R E 7 LncRNA TNRC6C-AS1 facilitated cell proliferation and inhibits cell apoptosis and autophagy by promoting the methylation of STK4 promoter region via the Hippo signalling pathway. Over-expressed TNRC6C-AS1 suppressed STK4 expression by promoting its methylation through methyltransferase recruitment. The protein levels of MST1 and LATS1/2 as well as the extent of YAP phosphorylation were reduced, thus activating YAP in thyroid carcinoma (TC) cells and promoting its nuclear translocation. As a result, the proliferation of TC cells was promoted while the apoptosis and autophagy of TC cells were inhibited