Identification of CCT3 as a prognostic factor and correlates with cell survival and invasion of head and neck squamous cell carcinoma

Abstract Background: Recurrent locally advanced or metastatic head and neck squamous cell carcinoma (HNSCC) is associated with dismal prognosis because of its highly invasive behavior and resistance to conventional intensive chemotherapy. The identification of effective markers for early diagnosis and prognosis is important for reducing mortality and ensuring that therapy for HNSCC is effective. Chaperonin-containing TCP-1 3 (CCT3) folds cancer-related proteins to control carcinogenesis. The prognostic value and growth association of CCT3 and HNSCC remain unknown. Methods: The GEO, Oncomine and UALCAN databases were used to examine CCT3 expression in HNSCC. A few clinical HNSCC samples with normal tissues were used to detect CCT3 expression by using immunohistochemistry method. The TCGA-HNSC dataset was used to evaluate the association between expression of CCT3 and prognosis. The molecular mechanism was investigated with gene set enrichment analysis (GSEA). CCK-8 and wound healing assays were used to detect cell growth and invasion of HNSCC, respectively. Results: CCT3 expression was significantly up-regulated in HNSCC in both mRNA and protein levels. In addition, up-regulated CCT3 expression was associated with various clinicopathological parameters. High expression of CCT3 was significantly correlated with inferior survival of HNSCC patients. Knockdown of CCT3 significantly inhibited cell growth and invasion of HNSCC cell lines. GSEA analysis indicated that CCT3 was closely correlated with tumor-related signaling pathways and HNSCC cell survival. Conclusion: Our findings suggest that CCT3 is a biomarker of poor prognosis and related to the process of HNSCC.


Introduction
Head and neck cancer (HNC) is a common malignant worldwide, and causes more than 600,000 new cases every year (1). Head and neck squamous cell carcinoma (HNSCC) accounts for the major pathological type of HNC (2). Although rapid advances in methods of diagnosis and treatment, the prognosis of patients with HNSCC remains not optimistic (3). Therefore, it is urgent to identify prognostic biomarkers for precise molecular diagnosis and potential targeted drug treatment in the future.
Chaperonins are kinds of proteins that help to fold stress-denatured polypeptide chains (4). Two groups of chaperonins are defined: group I includes Heat shock protein 60 (HSP60) or GroEL in bacteria, and group II are Chaperonin-containing TCP-1 (CCT). CCT consists of eight distinct subunits, named CCT1-CCT8 (5,6).
CCT3 is widely studied in various cancers (10)(11)(12)(13)(14)(15)(16). The expression of CCT3 is increased both at mRNA and protein level in hepatocellular carcinoma (HCC) tissues than those in non-HCC tissues, and CCT3 involves in carcinogenesis and development of HCC and has prognostic indication in HCC (10,11). Similar results are found in gastric cancer (12). In addition, knockdown of CCT3 decreased the cell viability of gastric cancer cells and accounted for inhibited expression of cell division cycle 42(cdc42), mitogen-activated protein kinase 7, cyclin D3 and upregulated of Downloaded from http://portlandpress.com/bioscirep/article-pdf/doi/10.1042/BSR20211137/920454/bsr-2021-1137.pdf by guest on 15 September 2021 Bioscience Reports. This is an Accepted Manuscript. You are encouraged to use the Version of Record that, when published, will replace this version. The most up-to-date-version is available at https://doi.org/10.1042/BSR20211137 5 cyclin-dependent kinase 2 and 6 (12). Silencing CCT3 inhibited the proliferation, stopped cell cycle and induced apoptosis in papillary thyroid carcinoma cells (13). In breast cancer, Xu et al. have reported that CCT3 knockdown significantly inhibited activity of NF-κB and decreased the proliferation and metastasis ability of breast cancer cells (14,15). Recently, a study performed a bioinformatic analysis focusing on the relationship between CCTs and HNSCC which indicated that CCT3 might be a potential biomarker for prognosis of HNSCC patients (16). Hence, we further explored the association between CCT3 and HNSCC.
In the present study, we have evaluated the relationship between CCT3 and HNSCC using analysis of comprehensive bioinformatics methods and basic experiments. CCT3 was overexpressed in tumor tissues of HNSCC than corresponding normal tissue. Increased CCT3 was significantly associated with process of HNSCC and suggested a poor prognosis in patients. Importantly, knockdown of CCT3 inhibited cell growth and cell migration of HNSCC cells. 7 The cell lysates of SCC25 and CAL27 cell lines were collected at 80-90% density. Twenty micrograms of total protein were fractionated with 10% SDS-page gel and transferred to 0.2 μm PVDF membrane (Biorad) at 100 V for 80 min. After being blocked with PBS containing 0.1% Tween 20 and 5% low-fat milk for 1 h at room temperature, the PVDF membrane was incubated with primary antibody overnight at 4 °C. The next day, after being washed 3 times with TBST, the membrane was cultured with secondary antibody for 1 h at room temperature, and subjected into ECL developing system. We used the rabbit polyclonal CCT3 antibody (cat no. ab244288, 1:1000, Abcam, UK) and mouse monoclonal anti-GAPDH (cat no. ab8245, 1:10000, Abcam, UK).

Cell viability
After transfection with siRNA, 1 x 10 4 SCC25 and CAL27 cell lines were collected and seeded in 96-well plates. Then cells were incubated with complete medium for 24 or 48 h. Ten micrograms of CCK-8 reagent (MedChemExpress) were added into each well. Cells were incubated in 37oC for 30 min and subjected to read OD value within 450 nm absorbancy.

Wound Healing
SCC25 and CAL27 were seeded and transfected in 12-well plate at 1 ml/well medium. After grown 80-90% density, cells were washed by PBS and were scratched HNSCC samples with clinical TNM information. These datasets were normalized through RMA method. UALCAN provides systematical analysis of TCGA gene expression data (http://ualcan.path.uab.edu), which allows clients performing in-depth and personalized analyzing. The expression of CCT3 was determined in the "TCGA-HNSC" dataset and with the "Expression Analysis" module. Furthermore, the mRNA expression of CCT3 between HNSCC and normal samples based on diverse clinicopathological parameters was also analyzed using the UALCAN database. ONCOMINE (www.oncomine.org) is currently the largest oncogene chip database which contains 715 gene expression data sets and contains independent integrated data mining function. Detailed expression of CCT3 in different primary sites of HNSCC was investigated using ONCOMINE. computer. The hazard ratio (HR) with 95% confidence intervals and log-rank P-value were obtained to evaluate the significance of prognosis in HNSCC patients.

The interaction network and co-expression analysis of CCT3.
To construct the protein-protein interaction (PPI) network of CCT3, GeneMANIA (http://www.genemania.org) and STRING (https://string-db.org/) was used. GeneMANIA provides predication of protein interaction and develops an interactive functional-association network which contains a list of genes with similar functions. STRING provides PPI network analysis including both certified and predicated links. CCT3 (protein name) and Homo sapiens (organism) were chosen.
GEPIA database is a comprehensive web-portal which allows users to perform a single or multiple gene analysis based on TCGA datasets. We explore the associations of CCT3 and other key candidate genes using the "Correlation Analysis" module.

Gene Set Enrichment Analysis (GSEA) of CCT3.
GSEA (http://www.broad.mit.edu/gsea)was performed to annotate the Hallmark and KEGG effector gene sets associated with the mRNA expression of CCT3 in the TCGA-HNSC dataset. The FDR<0.25 and P<0.05 was considered as significant.

Identification of essential role of CCT3 for cell survival.
Project Achilles comprehensively scores and identifies total 18000 genes which whether essential for cell survival in numbers of characterized cancer cell lines. The original data of CERES score for HNSCC cell lines were obtained from Depmap portal (https://depmap.org/portal). The CERES score approaches to 0 means when the gene was removing with CRISPR-Cas9 system, the cell growth has not been inhibited significantly. When the score less than -1, it means that the gene is essential for cell survival in this cancer cell line.

Statistical analysis.
The hazard ratio (HR) with logrank P value was used to evaluate the significance of survival. Spearman's correction was used to assess the association of gene expression and the strength of the correlation. Student's t test was used to evaluate the significance between two groups and triple repetitive experiments. The chi-square test was used to detect the significance of expression of CCT3 and clinicopathologic features. results were considered as statistically significance at *p < 0.05, **p < 0.01, and ***p < 0.001.

Expression of CCT3 in HNSCC.
To determine the expression of CCT3 in various cancers, TCGA and Oncomine databases were explored. As shown in Figure 1A, the mRNA level of and HNSCC was subjected into subsequent exploration. The CCT3 expression of TCGA-HNSC in normal and tumor tissue was presented in Figure 1C. As shown in Figure 1D and 1E, the expression of CCT3 in tongue squamous cell carcinoma (TSCC, Talbot lung and Estilo Head-Neck, Oncomine) were all significantly upregulated than in normal tongue tissues. To further confirm these results, one paired (GSE13398) and one non-paired (GSE29330) GEO datasets were explored. As shown in Figure 1F and 1G, CCT3 expression was significantly increased in tumor tissues than in normal oral samples. We further investigated the protein expression of CCT3 in HNSCC using IHC assay. Total 47 tumor samples and 15 normal oral tissues were analyzed. The expression of CCT3 was mainly accumulated in cytoplasm. The protein expression of CCT3 was significantly higher in OSCC than in normal epithelial tissues ( Figure 1H).
These results suggested that CCT3 was highly expressed in HNSCC than in associated normal tissues.

The prognostic analysis of CCT3 in HNSCC.
To explore whether CCT3 is associated with the prognosis of HNSCC patients, the OS of TCGA-HNSC dataset was investigated by using Kaplan-Meier plotter. As shown in Figure

Association of CCT3 expression and clinicopathological characteristics.
Since CCT3 expression was significantly increased in HNSCC tissues, we further explore the expression profiles of CCT3 based on clinicopathological parameters. The TCGA-HNSC datasets was analyzed by using the UALCAN database. Regarding the tumor grade, significant upregulation of CCT3 expression was found in grade 2 and grade 3 patients than in grade 1 patients ( Figure 3A). In addition, the expression of CCT3 was upregulated in stage 2 and stage 4 of HNSCC than in stage 1 ( Figure 3B). The CCT3 expression was significantly increased in African-american than in Caucasian and Asian ( Figure 3C). Regarding the nodal metastasis status, the mRNA expression of CCT3 was significantly upregulated in N3 patients with HNSCC than in N0 ( Figure 3D). Compared with HPV+ patients, CCT3 was highly expressed in HPV-patients ( Figure 3E). In addition, the CCT3 expression was upregulated in TP53-mutat patients than in TP53-normal patients ( Figure 3F).
Expression of CCT3 was not associated with patient's age ( Figure 3G) and gender ( Figure 3H). Because CCT3 was associated with nodal metastasis, the expression of CCT3 was investigated in GSE136037 dataset. The results showed that upregulated CCT3 was found in metastatic tumor tissues compared with primary tumor tissues ( Figure 3I). These results suggested that CCT3 may be associated with more serious process of illness. Since CCT3 was associated with the progress of HNSCC, we investigated the CERES dependence scores of HNSCC cell lines to determine the importance of CCT3 for survival of tumor cells. As shown in Figure 4A, when removing the expression of CCT3 genes with CRISPR-Cas9 system, total 21 HNSCC cell lines got the scores less than -1, which suggested that CCT3 plays a critical role for HNSCC cell survival. To further confirm these results, the sham siRNA or siRNA-CCT3 was transfected to SCC25 or CAL27 cell lines. As shown in Figure 4B, CCT3 was knock-down in both SCC25 and CAL27 cell lines. After transfected with siRNA, cells were subjected to a CCK-8 assay to detect cell viabilities at 24 h or 48 h. As shown in Figure 4C, knock-down of CCT3 significantly inhibited cell viability of SCC25 cells at 24 h checkpoint. In addition, downregulated CCT3 decreased cell viabilities of SCC25 and CAL27 cells after 48 h transfection ( Figure 4C and 4D). Then, a wound healing assay was performed to detect the cell invasion affected by CCT3. As shown in Figure 4E, after being transfected with siRNA-CCT3, the wound sizes were significantly decreased in both SCC25 and CAL27 cell lines. These results suggested that CCT3 was important for cell growth and invasion of HNSCC cells.

The Gene Set Enrichment Analysis of CCT3 in HNSCC.
To evaluate the possible mechanism of CCT3 on HNSCC, GSEA with the annotation of Hallmark and KEGG gene sets was performed. Total 43 and 131 important pathways were significantly affected by high expression of CCT3 in Hallmark and KEGG analysis, respectively (Supplemental Table 1  UBIQUITIN_MEDIATED_PROTEOLYSIS) were shown.

Identification of key candidate genes from the CCT3 interaction network.
To explore mechanisms of CCT3 in HNSCC and analyzed the function of these genes, a gene-gene interaction network for CCT3 was constructed using the GeneMANIA database. Total 20 nodes surrounding CCT3 represented genes that were significantly associated with CCT3 ( Figure 6A). The tightest corrected 5 genes were TCP1, CCT2, CCT4, PFDN2 and CCT6A. Additional functional analysis suggested that the proteins encoded by these genes were dramatically related with the following terms: 'De novo' posttranslational protein folding, 'De novo' protein folding, Protein folding, Unfolded protein binding, Chaperone-mediated protein complex assembly, Microtubule, and Cellular protein complex assembly. Then we constructed a PPI network using the STRING database to further explore the function of CCT3. A total of 10 CCT3-interacting proteins were included in the PPI network  Figure 6B). Importantly, eight common hub genes were identified from the GeneMANIA and STRING databases: TCP1, CCT8, CCT7, CCT6B, CCT6A, CCT5, CCT4 and CCT2. Co-expression analysis between CCT3 and these five interacting proteins were performed with the GEPIA database. As shown in Figure 6C, the expression of CCT3 was strongly correlated with that these genes except for CCT6B in HNSCC.

Discussion
In the present study, we conducted a comprehensive bioinformatics analysis and clinical sample assay to indicate that the expression of CCT3 was higher in HNSCC tissues than in corresponding normal tissues both in mRNA and protein levels. Overexpressed CCT3 in HNSCC tissues suggested poorer prognosis than patients with low-expressed CCT3. Importantly, knockdown of CCT3 leaded to inhibition of growth and invasion in HNSCC cell lines. Above all, CCT3 plays an important role in the progress of HNSCC.
The abnormal expression of CCT3 has been proved to influent the migration of tumor cells and the prognosis of cancer patients in previous researches. High expression of CCT3 was found in some cancers, such as hepatocellular carcinoma, multiple myeloma, colorectal cancer, liver cancer and gastric cancer (12,(19)(20)(21)(22), and increased expression of CCT3 was not only indicate a poor prognosis in patients with hepatocellular carcinoma, but also correlated with lymph-node metastasis of gastric cancer (23,24). Similar with these studies, our analysis suggested that overexpressed CCT3 was found in HNSCC tissues, and high level CCT3 indicated an inferior prognosis of patients with HNSCC. The mRNA expression of CCT3 was significantly upregulated in patients with more nodal metastasis status. proliferation of various cancer cells, such as papillary thyroid carcinoma, gastric carcinoma, breast cancer and hepatocellular carcinoma (12)(13)(14)27). In our study, the results of CERES dependence scores suggested that CCT3 was important for cell survival of HNSCC. In addition, knockdown of CCT3 with a special siRNA in HNSCC cell lines leaded to the growth suppression of cancer cells, which is consistent with studies in other cancers (12)(13)(14)27). Importantly, higher expression of CCT3 was found in nodal metastasis patients. In addition, the expression of CCT3 in metastatic lesions was higher than in primary tumor tissues. These results suggested that CCT3 may be associated with metastatic capacity of HNSCC. Consistently, the wound healing assay indicated the inhibition of cell invasion when knockdown of CCT3 expression. Above all, CCT3 might be a potential drug target in HNSCC.
In clinical subgroup analysis, expression of CCT3 was associated with higher cancer stages and tumor grades. Regarding mechanisms, CCT3 affects the progression of HCC by activating signal transducer and activator of transcription 3 (STAT3) (10,28). STAT3 is the major factor in JAK-STAT3 pathway signaling, which plays an important role in many aspects of tumorigenesis (29). The activation of STAT dimers in nucleus can be affected by mitogen-activated protein kinase (MAPK), AKT/mammalian target of rapamycin (mTOR) and JAK (30), and a recent study disclosed that mTORC, which is multi-protein signaling complex of mTOR, assembly and signaling can be affect by eukaryotic chaperonin CCT(31). Qian et al. has suggested that CCT3 expression was associated with JAK-STAT3 rather than mTOR pathway by KEGG and GSEA analysis (21). However, in the present study, our results