Coiled-Coil Domain-Containing Protein 45 Is a Potential Prognostic Biomarker and Is Associated with Immune Cell Enrichment of Hepatocellular Carcinoma

Objective The role of coiled-coil domain-containing protein 45 (CCDC45) in the development of hepatocellular carcinoma (HCC) has not been reported. The present study is aimed at investigating the expression and prognosis of CCDC45 in HCC and its relevance to immune infiltration. Methods We conducted CCDC45 expression analysis using The Cancer Genome Atlas (TCGA) tumor database, the Human Protein Atlas (HPA) database, and the Tumor Immunological Evaluation Resource (TIMER). We used the University of Alabama at Birmingham Cancer data analysis Portal (UALCAN) database to show the correlation of CCDC45 with clinical features. We examined the prognostic impact of CCDC45 expression levels on HCC patients with the Kaplan-Meier mapper database. Genes coexpressed with CCDC45 and its regulators were also identified using LinkedOmics. The enriched Gene Ontology (GO) categories and associated signaling pathways were estimated using GO, Kyoto Encyclopedia of Genes and Genomes (KEGG), and Gene Set Enrichment Assay (GSEA) pathway data. Correlations between CCDC45 and cancer immune infiltration was analyzed through the TIMER and an integrated repository portal for Tumor-Immune System Interactions (TISIDB) databases. Results The expression of CCDC45 was elevated in HCC tissues compared to adjacent liver tissues, and overexpression of CCDC45 was significantly correlated with tumor stage. Furthermore, HCC patients with CCDC45 overexpression had a shorter overall survival (OS). Functional network analysis indicated that CCDC45 was involved in homologous recombination, spliceosome, and DNA replication. Interestingly, CCDC45 expression was positively correlated with the level of immune cell infiltration. Conclusions CCDC45 is associated with prognosis and immune infiltration of HCC and may be a potential therapeutic target for HCC.


Introduction
Hepatocellular carcinoma (HCC) is the most common primary liver tumor, but the molecular mechanism of HCC development is not well understood [1,2]. Despite substantial advances in targeted therapies for HCC (e.g., tyrosine kinase inhibitors, such as sorafenib and lenvatinib), previous studies have shown that most patients with HCC developed acquired resistance to targeted agents (e.g., PD-1) in a relatively short period of time [3][4][5]. HCC is a highly heterogeneous tumor [6]. Interactions with the HCC cell and tumor microenvironment have provided new directions to investigate the molecular mechanisms by which HCC develops and metastasizes; however, the molecular drivers of HCC carcinogenesis remain unclear [7,8]. In the current management of HCC, there is a lack of reliable early biological markers [9].
Thus, it is essential to identify reliable molecular markers for the early diagnosis and treatment of HCC patients.
Coiled-coil domains constitute a family of proteins containing coiled-coil domains (CCDCs), which are involved in many functions in cell growth and development, such as regulation of gene expression and drug sensitivity [10][11][12][13][14][15]. The abnormal expression of CCDC65 and CCDC88 proteins has been shown to significantly affect the malignant progression of gastric and pancreatic cancers, providing a theoretical basis for identifying oncology drug targets. Coiled-coilcontaining domain 45 (CCDC45) is located on chromosome 17q23.3 of the human genome and encodes a protein of 1200 amino acids. CCDC45 is also known as centrosomal protein 95 (CEP95). Increasing evidence confirms the dysregulated expression of CCDC family genes in tumors, but the expression and clinical significance of CCDC45 in tumors, including HCC, remains uncertain.
In the present study, we utilized several biological databases and multiple experimental approaches to analyze the differential expression of CCDC45 in HCC and its relationship with clinicopathological parameters. Furthermore, we applied Gene Ontology (GO) annotation, Kyoto Gene and Genome Encyclopedia (KEGG), and Gene Set Enrichment Analysis (GSEA) to explore the potential functions of CCDC45 in HCC. In addition, immune infiltration of HCC cells is an important cause of HCC development; hence, we also explored the relevance of immune cells in the immune microenvironment of HCC to the expression of CCDC45. Our findings suggested that CCDC45 may be a potential therapeutic target and a reliable prognostic indicator for HCC.  Expression levels of CCDC45 in unpaired HCC tissues in TCGA database (c) expression levels of CCDC45 in paired HCC tissues in TCGA database. (d) Comparison of CCDC45 expression levels in normal liver tissues and HCC tissues based on the HPA database. * P < 0:05; * * P < 0:01; * * * P < 0:001; ns, not significant, P > 0:05.

Disease Markers
browser, censoring the data cohort used as survival analysis and clinicopathological parameter analysis due to missing data. The transcriptome data were organized into gene expression matrix files using the RSEM program (v1.3.3), and the expression data of CCDC45 in each group of samples were extracted using the R program (v3.6.0). P < 0:05 was considered statistically significant.

UALCAN Database.
UALCAN is a website that allows online analysis of clinicopathological features of TCGA data, and in this study, we used UALCAN to compare the correlation between CCDC45 and different clinicopathological features, such as tumor grade, stage, patient age, race, and molecular subtype of tumor with gene expression [16]. P < 0:05 was considered statistically significant.       [17].
In the present study, we compared the protein expression of CCDC45 in the liver and HCC tissues from the HPA database.

LinkedOmics Database.
The LinkedOmics database is a multiomics site that allows analysis of all tumors in TCGA dataset [18]. In the present study, we queried differentially expressed genes associated with CCDC45 in HCC through the LinkFinder module. Outcomes were analyzed by Pearson correlation coefficients, and these different genes were visualized using volcano plots and heatmaps. To further explore the biological functions that CCDC45 may be involved in, these genes were subjected to GO analysis and KEGG pathway enrichment analysis.

TISIDB Database Analysis.
TISIDB is an online analysis of various tumor and immune interactions, aiding in the speculation of the immune effects of immunotherapy [19].
In the present study, the association of CCDC45 expression with immunomodulators and chemokines were analyzed by the TISIDB database. P < 0:05 was considered statistically significant.

Tumor Immune Estimation Resource (TIMER) Analysis.
TIMER is an online website for the analysis of immune infiltration in different tumors [20,21]. To investigate the relationship between immunological infiltration of CCDC45 and HCC, we estimated the abundance of immune cells, including B cells, CD8 + T cells, CD4 + T cells, macrophages, natural killer (NK) cells, and dendritic cells (DCs), through the TIMER algorithm.

GSEA of CCDC45 Expression
Phenotype. Based on the expression of CCDC45, all HCC patients were divided into two groups according to high and low expression of CCDC45 using the UCSC Xena platform to download data from TCGA-HCC, and each analysis was repeated 1000 times to align all genes and analyze the enrichment pathways in the CCDC45 high expression group by GSEA [22,23]. Data were considered statistically significant if P < 0:05 and the data were below the FDR-q of 0.25.

Disease Markers
Response to ischemia Response to pH Response to type I interferon Benzene-containing compound metabolic process Cellular modified amino acid biosynthetic process Tetrapyrrole metabolic process Sulfur compound metabolic process Cytolysis Cellular aldehyde metabolic process Vitamin metabolic process Isoprenoid metabolic process Nucleoside bisphosphate metabolic process Mitochondrial respiratory chain complex assembly Lipid catabolic process Platelet degranulation Cellular modified amino acid metabolic process Cellular amino acid metabolic process NADH dehydrogenase complex assembly Dicarboxylic acid metabolic process Peroxisome organization Humoral immune response Peroxisomal transport Antibiotic metabolic process Acute inflammatory response     10 Disease Markers 2.9. Western Blot Analysis. Cells were lysed sufficiently in protein lysis buffer (radioimmunoprecipitation assay buffer, RIPA) and further fragmented under sonication conditions. After centrifugation, supernatants were collected, and protein levels were determined by a bicinchoninic acid (BCA) protein quantification assay [25]. Proteins were separated on 10% sodium dodecyl-sulfate (SDS) gels and transferred to polyvinylidene fluoride (PVDF) membranes. The membranes were incubated with primary antibodies against CCDC45 (ab230305, Abcam, 1 : 1000) and GAPDH (60004-     CD27  CD276  CD28  CD40  CD40LG  CD48  CD70  CD80  CD86  CXCL12  CXCR4  ENTPD1  HHLA2  ICOS  ICOSLG  IL2RA   IL6R  KLRC1  KLRK1  LTA  MICB  NT5E  PVR  RAET1E  TMEM173  TMIGD2  TNFRSF13B  TNFRSF13C  TNFRSF14  TNFRSF17  TNFRSF18  TNFRSF25  TNFRSF4  TNFRSF8  TNFRSF9  TNFSF13   TNFSF14  TNFSF15  TNFSF18  TNFSF4  TNFSF9  ULBP1 TNFSF13B (d)  CESC  CHOL  COAD  ESCA  GBM  HNSC  KICH  KIRC  KIRP  LGG  LIHC  LUAD  LUSC  MESO  OV  PAAD  PCPG  PRAD  READ  SARC  SKCM  STAD  TGCT  THCA  UCEC  UCS  UVM   1 -0  14 Disease Markers Physician. All scores were determined based on staining intensity and range. The intensity of staining was scored based on the following scale: 0, negative expression; 1, weak expression; 2, moderate expression; and 3, strong expression. The staining range was scored based on the following 4-point scale: 0, 0% staining area; 1, staining range of 1% to 24%; 2, staining range of 25% to 49%; 3, staining range of 50% to 75%; and 4, staining range of 75% to 100%. The results of the two separate scores were multiplied to obtain the final score. Patients were included in the low expression CCDC45 group if their total score was <6, while those with a total score ≥ 6 were included in the high expression CCDC45 group.
2.11. Survival Analysis. Kaplan-Meier analysis is an online site created by Oncomir to analyze the survival of various genes in relation to tumors. In the present study, we divided CCDC45 into two groups according to their median expression and examined the correlation between CCDC45 in HCC and patient prognosis by log-rank test to determine whether the survival curves of the two groups were statistically different. We also used univariate and multivariate COX regression analyses to validate the relationship between CCDC45 and HCC prognosis. SPSS 22.0 were used to determine significance (P) value, and P < 0:05 was considered statistically significant.

CCDC45 Expression Is Increased in HCC.
We used the TIMER2.0 database to explore the expression levels of CCDC45 in normal and various cancer tissues. The results showed that CCDC45 was highly expressed in most tumors, including liver hepatocellular carcinoma (LIHC), and only a few tumors had low CCDC45 expression (Figure 1(a)). A PUBMED database search indicated that the CCDC45 gene has not been reported in tumors. Thus, we focused on CCDC45 in the present study and downloaded TCGA database of HCC-related mRNA expression data to analyze the expression of CCDC45 in HCC. The results demonstrated that CCDC45 was significantly overexpressed in unpaired samples (Figure 1(b)) and paired samples (Figure 1(c)) of HCC, which agreed with the previous pancancer analysis. In addition, CCDC45 protein levels were also higher in HCC tissues than in normal liver tissues in the HPA database (Figure 1(d)). Thus, these findings indicated that CCDC45 expression is upregulated in HCC tissues compared to adjacent liver tissues.    Figure 3(a)). Similar results were obtained for progression-free survival (PFS) (HR = 1:34, 95% CI: 1.00-1.79, P = 0:05, Figure 3(b)). Interestingly, CCDC45 expression showed significant differences with the prognosis of HCC patients, and higher CCDC45 expression indicated worse prognosis (Figures 3(c) and 3(d)). The above results suggested that CCDC45 may be a marker that influences disease progression in HCC. Additionally, we further investigated the predictive accuracy and risk score of the effect of CCDC45 on HCC survival using receiver operating characteristic (ROC) curves. CCDC45 was found to be a good predictor of the prognosis of HCC patients at 1, 3, and 5 years with area under the curve (AUC) values of 0.672, 0.597, and 0.565, respectively (Figure 3(e)). Thus, these results indicated that CCDC45 expression is associated with disease progression in HCC patients.

Analysis of CCDC45 Coexpressed Genes in HCC Using
GO and KEGG Analyses. To investigate the biological significance of CCDC45 in HCC, we analyzed the coexpression network of CCDC45 in the LIHC cohort through the Linkedomics database. As shown in Figure 4(a) 5,328 genes (red dots) were significantly positively associated with CCDC45, while 2,460 genes (green dots) were significantly negatively associated with ASF1B (false discovery rate, FDR < 0:01). We identified the top 50 genes positively associated with CCDC45 expression and the top 50 genes negatively associated with CCDC45 expression, and they were visualized using a heatmap (Figures 4(b) and 4(c)). The genes most positively associated with CCDC45 included TLK2, DDX42, and POLG2, while those negatively associated with CCDC45 included PINK1, AIFM1, and ACAT1. These

17
Disease Markers genes were subjected to further GO and KEGG enrichment analyses. GO analysis indicated that these genes were enriched in the following GO terms: biological processes, including chromosome segregation, DNA replication, and cell cycle G2/M phase transition (Figure 4(d)); cellular components, including condensed chromosomes, chromosomal regions, spliceosomal complexes, and other related (Figure 4(e)); and molecular functions, including helicase activity, catalytic activity, acting on DNA, and histone binding (Figure 4(f)). KEGG analysis indicated that these coexpressed genes were mainly involved in homologous recombination, spliceosome, and DNA replication ( Figure 4(d)). The above results indicated that the coexpression of CCDC45 is closely related to the processes involved in the development of HCC.  Figure 5(a)). Intriguingly, high levels of CCDC45 expression significantly correlated with high levels of infiltration of most immune cells, including T cells, helper T cells, NK cells, neutrophils, aDCs, B cells, CD 8 + T cells, and regulatory T cells (P < 0:05) ( Figure 5(b)). The single-sample GSEA (ssGSEA) algorithm was used to determine the correlation between CCDC45 and immune cell enrichment, and the results showed that CCDC45 expression was closely associated with T cells, DCs, NK cells, and macrophages ( Figure 5(c)). In addition, CCDC45 was significantly associated with immunostimulants, such as C10orf54 These results suggested that CCDC45 may serve as an immunomodulatory factor in HCC.
3.6. GSEA of the Signaling Pathways Associated with CCDC45. To further determine the function of CCDC45, GSEA was performed. The results showed that high expression of CCDC45 may be involved with the following pathways: mitotic spindle, MYC target, E2F target, DNA repair, apoptosis, unfolded protein response, P53 pathway, PI3K/AKT/mTOR pathway, and Wnt-β-catenin pathway ( Figure 6).

CCDC45 Is Highly Expressed in HCC.
For the verification of CCDC45 expression in HCC, we investigated the performance of three HCC cell lines and one immortalized hepatic epithelial cell line (LO2) by qRT-PCR and Western blot analyses to examine the expression level of CCDC45 in cells. Compared to LO2 cells, CCDC45 expression levels were significantly upregulated in HCC cells at both the mRNA (Figure 7(a)) and protein (Figure 7(b)) levels. To further understand the relationship between CCDC45 expression in HCC and clinicopathological parameters of HCC patients, we utilized a tissue microarray containing 90 pairs of HCC tissues and paracancerous tissues for immunohistochemical staining, and the results are shown in Figure 7(c). CCDC45 expression was significantly higher in HCC tissues compared to adjacent paracancerous tissues. Survival analysis showed that high CCDC45 expression was associated with poor prognosis in HCC patients, and patients with low expression of CCDC45 tended to have longer survival times (Figure 7(d)). We then further assessed the relationship between CCDC45 and the clinicopathological features of HCC by a Chi-square test based on the results of immunohistochemistry (Table 1). CCDC45 expression was associated with Edmondson-Steiner grade (P = 0:018). Univariate COX regression analysis showed that American Joint Committee on Cancer (AJCC) stage (P = 0:014), GGT (P = 0:044), Edmondson-Steiner grade (P < 0:001), tumor size (P = 0:041), and CCDC45 expression (P < 0:001) were associated. Intriguingly, multivariate Cox regression analysis indicated that Edmondson-Steiner grade (P = 0:003) and CCDC45 expression (P < 0:001) were strongly correlated ( Table 2). Thus, these findings suggested that CCDC45 is an independent factor for the prognosis of HCC patients.

Discussion
With recent advances in precision medicine and immunotherapy, a major breakthrough has been made for the treatment of HCC, but HCC still presents many challenges in the therapeutic field due to its highly heterogeneous nature [26][27][28]. In addition to molecular pathology research, it is of great interest to discover molecular markers of diagnostic significance in HCC through various bioinformatics tools [8,[29][30][31][32].
Investigation of the expression of CCDC family genes in various tumors has been the subject of many studies in recent years [33,34]. Numerous studies have shown abnormal expression of the CCDC family genes in various tumors, including nonsmall cell lung cancer [35], liver cancer [11], renal cell carcinoma [36], gastric cancer [10], breast cancer [37], and other malignancies. However, CCDC45 has not been reported in HCC. In the present study, we first reviewed databases and found that CCDC45 was upregulated in HCC tissues and that high CCDC45 expression in HCC often predicted a poor prognosis. Moreover, ROC curve analysis suggested that CCDC45 may have predictive value as a biomarker for HCC. Furthermore, we found that CCDC45 overexpression was closely associated with age, gender, race, histological grade, pathological stage, and distant metastasis in HCC patients. The present study is the first to explore CCDC45 expression in HCC by querying multiple bioinformatics databases and combining them with experiments. In the present study, survival analysis indicated that the expression of CCDC45 in HCC was significantly associated with shorter survival, which was supported by several other databases. Multivariate Cox analysis confirmed that CCDC45 is an independent risk factor for HCC patients.
To elucidate the biological role of CCDC45 in HCC, we queried the LinkedOmics database and found that CCDC45 not only had a significant impact on the prognosis of HCC patients but that most of the genes that coexpressed with CCDC45 in HCC also significantly correlated with the prognosis of HCC patients. Furthermore, these coexpressed genes were involved in homologous recombination, spliceosome, and DNA replication. Moreover, GSEA showed that high expression of CCDC45 may be involved in the following pathways: mitotic spindle, MYC target, E2F target, DNA repair, apoptosis, unfolded protein response, P53 pathway, PI3K/ AKT/mTOR pathway, and Wnt-β-catenin pathway.
HCC cells are involved in a complex tumor microenvironment, including extracellular matrix, stromal cells, and

20
Disease Markers immune cells, which together form the immune microenvironment in which HCC cells grow and metastasize [38][39][40][41]. To further examine the relationship and potential mechanisms between CCDC45 and tumor immune microenvironment, we conducted a correlation analysis with immune-related genes. CCDC45 showed positive correlation with the expression of many immune related cells in HCC, including T cells, helper T cells, NK cells, neutrophils, aDCs, B cells, CD8 + T cells, and regulatory T cells. Furthermore, poor prognosis of the CCDC45 overexpression group was associated with an overregulation of immune-related suppressor genes and chemokine receptors [42]. Taken together, these findings suggested that targeting CCDC45 may be a potential strategy for immune checkpoint therapy. Although our multidimensional analysis of CCDC45 and cross-validation using multiple databases provide insight into the association between CCDC45 and HCC to some extent, there were certain limitations in the present study. First, the lack of specificity of the sequencing data due to analyzing different database sources may have introduced sample bias. Moreover, there was a lack of in vitro and in vivo experiments to validate the potential functional effects of CCDC45 on HCC. Thus, future studies focusing on the mechanistic effects of CCDC45 on HCC are required. Although our data suggested that CCDC45 expression is closely associated with immune cell infiltration, the present study lacked direct evidence showing that CCDC45 affects HCC patients through its involvement in immune cell infiltration. Therefore, the mechanism of CCDC45 involvement in the development of HCC through immune regulation remains to be further investigated.

Conclusions
In summary, the present study demonstrated that CCDC45 is highly expressed in HCC tissues and that high expression of CCDC45 often predicts a poor prognosis. Moreover, the present findings indicated that CCDC45 is an independent prognostic factor for HCC and that the expression of CCDC45 is positively correlated with T cells, B cells, and NK cells. The present study provides new insights into the mechanisms associated with the infiltration of CCDC45 and immune cells in the tumor microenvironment of HCC. We also preliminarily investigated the biological processes associated with CCDC45, providing new directions for further typing studies and clinical applications.

Data Availability
The bioinformatics data from this study are available in an online database. Additional data are available from the authors upon request.