Increased Expression of CDCA2 in Glioma Predicts Poor Prognosis as Identified by Population-based Analysis

Background: The cell division cycle-associated protein 2 (CDCA2) was found to be a cell cycle-related protein. CDCA2 was previously reported to be associated with proliferation and migration in multiple cancer. We evaluated the role of CDCA2 in glioma using publicly available data from the Chinese Glioma Genome Atlas (CGGA) and The Cancer Genome Atlas (TCGA). Methods: The relationships between clinical characteristics and CDCA2 expression were analyzed using the Wilcoxon test or Kruskal-Wallis test. Clinicopathologic characteristics associated with overall survival (OS) were analyzed using Cox regression analysis and Kaplan-Meier method. Gene Set Enrichment Analysis (GSEA) was performed to sort the signaling pathway associated with the expression of CDCA2. Pearson correlation test was used to analyse the expression correlation between CDCA2 and well-known cell cycle-related genes. Results: Glioma with high CDCA2 expression was more prone to be associated with advanced malignancy clinical pathologic characteristics and worse prognosis than that with low CDCA2 expression in TCGA and CGGA dataset (P < 0.001). The multiple analysis revealed that CDCA2 was independently associated with OS (HR:1.396; [CI]:1.236 − 1.577, P < 0.001). GSEA showed that cell cycle checkpoint, cell cycle G1/S phase transition, DNA damage checkpoint and regulation of cell cycle arrest were enriched in CDCA2 high expression phenotype. Pearson correlation test revealed that CDCA2 was co-expression with well-known key cell cycle-related genes (CCNA2, CCNB1, CCNB2, CCNE1, CCNE2, CDK1, CDK2, CDK4, CDK6). Conclusion: High CDCA2 expression may be a potential prognosis molecular marker of poor survival in glioma. Cell cycle regulation pathway may be the key pathway regulated by CDCA2 in glioma. 95% between CDCA2 well-known cell cycle-related genes Receiver operating (ROC) curve and area (AUC) to evaluate the prediction accuracy of CDCA2 expression for 1-year, 3-year and 5-year

pathogenesis and development of glioma has become the focus of current research.
Cell division cycle-associated 2 (CDCA2) is a cell cycle-related nulear protein, which can bind to protein phosphatase 1 γ (PP1γ) and regulate DNA damage response in cell cycle [8,9]. The expression of CDCA2 was also found to be correlated with several well-known cell cycle-related genes, such as CDC2, CDC7, CDC23, cyclin, MCAK, MKI67a [10]. Recent reports indicated that CDCA2 was upregulated in colorectal cancer [11], esophageal squamous cell carcinoma [12], breast cancer [13] and might act as a prognostic factor. Previous literature reported that the expression of CDCA2 mRNA was higher in neuroblastoma than types of glioma [14]. However, the correlation between CDCA2 expression and the prognosis of glioma has not been study.
In the present study, we analyzed the expression level of CDCA2 to investigate its prognostic value in glioma. Furthermore, GSEA was performed to gain insight into the biological pathway in glioma pathogenesis related to CDCA2 regulatory network.
We demonstrated that high expression of CDCA2 might be an independent prognostic indicator associated with poor survival in glioma. GSEA showed that cell cycle regulation pathways were associated with CDCA2 high expression phenotype. CDCA2 expression was also positively correlated with the expressions of well-known key cell cycle regulation genes.

Source population
The gene expression data was downloaded from TCGA official website for low grade glioma (LGG) and glioblastoma multiforme(GBM) project and corresponding clinical information was downloaded from UCSC Xena (https://xenabrowser.net/). In CGGA database, the gene expression data and corresponding clinical information were downloaded from official website (http://www.cgga.org.cn/).
The gene expression levels displayed in CGGA and TCGA database had been standardized. Patients with unknown information of survival and essential clinical characteristics were excluded in this study.
The clinical information and gene expression profiles of patients with glioma in the CGGA dataset (n = 748) include age, gender, histology, the WHO grade, radiotherapy, chemotherapy, isocitrate dehydrogenase (IDH) mutation status, chromosome 1p/19q co-deletion status, recurrent status and vital status; in the TCGA dataset (n = 513), clinical information includes age, gender, histology, the WHO grade, radiotherapy, recurrent status and vital status, as shown in Table 1. survival. The goal of GSEA is to utilize a predefined group of genes (mainly by previous biological knowledge, published information about biochemical pathways or co-expression in previous experiments) for ranking the genes in accordance with the extent of differential expression within the 2 types of samples, then verifying that the predefined group of genes tend to occur toward the bottom or top in the sorting table [15]. To explore the differences in pathways as well as biological functions in the low-and high-expression sets of such prognostic CDCA2 gene, GSEA was used to explore potential pathway and GO analysis within the Molecular Signatures Database (MSigDB) and c5 (GO gene sets). The nominal p-value and normalized enrichment score (NES) were used to sort the pathway enriched in each phenotype. All statistical analyses were conducted using R statistical software (version 3.6.2). P < 0.05 was considered to indicate a statistically significant difference.

Patient characteristics and clinical pathologic features in TCGA and CGGA dataset
As shown in Table 1 Association with CDCA2 expression and clinicopathologic variables in TCGA and CGGA dataset from CGGA dataset. As shown in Fig. 1 (A-G), increased expression of CDCA2 significantly correlated with the old age, recurrent status, chemotherapy, WHO IV grade, IDH wild type, 1p/19q non-codeletion status and GBM histology type (all p-value < 0.05). The high expression level of CDCA2 was correlated with old age compared with younger age (Fig. 1A). CDCA2 expression level was higher in recurrent status (Fig. 1B). CDCA2 high expression level was positively correlated with glioma grade ( Fig. 1D and G). The expression levels of CDCA2 were higher in the IDH wild-type and 1p/19q non-codeletion status compared with the corresponding group ( Fig. 1E and F). In TCGA dataset, as shown in

CDCA2 involved in cell cycle regulation signaling pathway
To identify signaling pathway associated with CDCA2 expression that are differentially activated in patients with glioma, we conducted GSEA analysis between low and high CDCA2 expression data sets.
GSEA reveals significant differences (false discovery rate (FDR) < 0.05, NOM p-val < 0.05) in enrichment of MSigDB Collection (C5.all.v.7.1symbols and h.all.v.7.1symbols). We selected the significant enriched signaling pathways based on their NES ( Fig. 4 and Table 2). The Fig. 4 showed that CDCA2 high expression phenotype was associated with cell cycle checkpoint, cell cycle G1/S phase transition, DNA damage checkpoint and regulation of cell cycle arrest in GSEA analysis.

Discussion
Glioma is the most common and lethal type of brain tumor, which accounts for 46% of intracranial tumors [16]. With the development of sequencing technology, molecular factors for prognosis of glioma have been well studied, such as IDH mutations, 1p/19q co-deletion, ATRX mutation, TERT promoter mutation and PTEN loss [17][18][19][20]. Howerer, traditional therapeutic approaches, including surgery, radiotherapy, chemotherapy and targeted therapy do not achieve satisfactory results.
Genome differences in patients with glioma make different prognosis of patients. It is necessary for us to explore more prognostic markers to further understand the mechanism of glioma and predict prognosis of patients with glioma. The researches of cancer development revealed that CDCA2 was related to the occurrence and development in multiple cancers [11,12,21,22]. To our knowledge, the expression of CDCA2 and its potential prognostic impact on glioma has not been explored. The potential role of CDCA2 in glioma is the focus in our present study.
In the present study, bioinformatic analysis using high throughout RNA-sequencing data from TCGA and CGGA revealed that an increased expression of CDCA2 in glioma was associated with advanced clinical pathologic characteristics (old age, recurrent status, high grade and GBM sub-type), shorter survival time and poor prognosis. Further, multivariate Cox proportional hazards analysis showed that increased expression of CDCA2 was demonstrated to be an independent prognostic factor in the survival time of patients with glioma in CGGA dataset. To investigate the function of CDCA2 in glioma, we conducted GSEA using CGGA data. GSEA showed that cell regulation pathways were enriched in CDCA2 high expression phenotype. Pearson's correlation analysis also showed that increased expression of CDCA2 was positively correlated with well-known key cell regulation genes. This suggested that CDCA2 might serve as a potential marker of prognosis and therapeutic target in glioma.
Previous study reported that the expression of CDCA2 mRNA was higher in neuroblastoma than that of lower stage tumor [14]. CDCA2 was also over-expressed in other cancer types and associated with poor prognosis. Recent study demonstrated that CDCA2 promoted colorectal cancer cells proliferation by activating the AKT/CCND1 pathway in vitro and in vivo [11]. More studies were carried out to identify that CDCA could be an signature gene with prognostic value for luminal breast cancer, esophageal squamous cell carcinoma, bladder cancer, melanoma and Synovial sarcoma [12,13,21,23]. In this work, we also demonstrated that strong expression of CDCA2 in glioma was associated with advanced clinical pathologic characteristics and higher expression of CDCA2 was an an independent prognostic indicator associated with poor survival in glioma.
CDCA2 was first identified by Walker as a novel cell-cycle associated gene using mircoarray analyses of co-expression of the well-known cell-cycle genes[10]. Mulcahy and Lamond further revealed that CDCA2 binded to protein phosphatase 1 (PP1), responsible for the targeting of PP1 to chromatin in anaphase. CDCA2/PP1 complex was involved in cell cycle regulation and proliferation [24].Vagnarelli demonstrated that the CDCA2/PP1 complex was critical component of the chromatin reorganization machinery responsible for chromosome de-condensation at the transition from mitosis to G1 [25].
Moreover, the study showed that the complex modulated ATM activation, setting the threshold for checkpoint activation[9, 26,27]. CDCA2 was released from the chromatin at DNA damage sites, which presumably facilitated DNA damage response (DDR) activation [8]. DDR was activated in precancerous cells as a barrier to suppress cell proliferation, cancer progress and reduced in late-stage cancer cells. It was possible that strong expression of CDCA2 could result in desensitization of cells to DDR [28]. Depletion of CDCA2 re-actives the DDR and drive the cells into apoptotic pathway [29]. In our study, we observed that CDCA2 high expression phenotype was associated with cell cycle checkpoint, cell cycle G1/S phase transition, DNA damage checkpoint and regulation of cell cycle arrest in GSEA analysis. In OSCC cell line, depletion of CDCA2 caused a decrease in cell proliferation due to cell cycle arrest in G1 phase and down-regulation of CDK4, CDK6, Cyclin D1/E [22]. In our study, we also demonstrated that the expression of CDCA2 was positively correlated with expressions of well-known key cell cycle regulation genes (CCNA2, CCNB1, CCNB2, CCNE1, CCNE2, CDK1, CDK2, CDK4, CDK6). Cyclin family plays a pivotal part in cell cycle regulation and is involved in a range of biological processes [30,31]. CDKs, a family of proteins that are involved in the regulation of the cell cycle, are frequently over-expressed or mutated in cancer, and CDK2/4/6 inhibitors (CDK2/4/6i) have been developed to be relatively safe and effective cancer therapeutics [32,33]. From the work presented above, it appears that the CDCA2 may be a crucial functional factor in mitosis and a crucial hub for the regulation of chromatin organization and the maintenance of genome stability. Further work is needed to determine what is the molecular role of CDCA2 in glioma cell cycle regulation pathway and can it be a potential target therapy for glioma treatment? However, the prediction expression of protein using mRNA expression was far from perfect. The correlation between CDCA2 mRNA expression and CDCA2 protein expression could not be clearly identified because of the limitation in our study. Further study in patients with glioma is required.
In conclusion, high CDCA2 expression may be used as an independent prognostic molecular indicator of poor survival in patients with glioma. Moreover, the expression of CDCA2 was positively correlated with expressions of well-known key cell cycle regulation genes, and the cell cycle regulation pathway may be the key pathway regulated by CDCA2 in glioma. Further experimental validation required to prove the biological impact of CDCA2 in glioma.   Positive correlation between the expression of CDCA2 and cell cycle regulation genes.