GINS1/2/3/4 Upregulation Predicts Poor Prognosis in Human Sarcoma


 Background: GINS family was reported to be highly expressed in many tumors. However, the association of GINS family with human sarcoma remained unknown. This study was undertaken to explore the expression and prognostic value of GINS family in human sarcoma.Methods: In terms of the expression levels of mRNA for GINS family members, a particular contrast in various cancers, especially human sarcoma, was conducted through ONCOMINE and GEPIA and CCLE databases. Kaplan-Meier Plotter was used to identify the prognostic value of GINS family in sarcoma.Results: We discovered that the mRNA expression levels of GINS1, GINS2, GINS3, and GINS4 were all higher in the majority of tumor tissues than in normal samples, of course, including sarcoma. Through the Cancer Cell Line Encyclopedia (CCLE), all the four members (GINS1, GINS2, GINS3, GINS4) expression were observed in high levels in sarcoma cell lines. In Gene Expression Profiling Analysis (GEPIA) and Kaplan-Meier Plotter, our results indicated that the poor overall survival (OS), disease-free survival (DFS) and relapse free survival (RFS) were tightly associated with the increased expression of GINS genes. Conclusion: The four GINS family members are all the prognostic biomarkers for the prognosis of human sarcoma.


Background
Sarcomas consist of a group of rare solid tumors of mesenchymal cell origin with distinct clinical and pathological features. Generally, they are divided into two categories: soft tissue sarcoma and osteosarcoma. Sarcomas account for approximately 1% of all adult malignancies and 15% of pediatric malignancies. [1] Osteosarcoma accounts for 20% of primary malignant bone tumors, it is the most common primary malignant bone tumor in adolescents. 70%-80% of all the patients are 10-25 years old.
The annual incidence rate is 1-3 people per 1 million people. [2] As the pathogenesis of osteosarcoma is still unclear, the 5-year survival rate of patients with osteosarcoma still hovers between 60% and 70%. [3] Soft tissue sarcomas (STS) (including fat, muscle, nerve and nerve sheath, blood vessel and other connective tissues) is classi ed according to similar tissue morphology. The STS classi cation reported by the World Health Organization in 2013 includes more than 50 histological subtypes, liposarcoma (LPS), and Differentiated sarcoma (US), leiomyosarcoma (LMS), myxo brosarcoma (MFS) and synovial sarcoma (SS) are the most common types in adults. [4] Despite the development of surgery and adjuvant therapy, the treatment of sarcoma has improved [5], but the long-term survival rate of patients with highgrade sarcoma is still very low. [6] More than 50% of high-risk sarcoma patients will metastasize and die. Therefore, it is urgent to further explore the molecular mechanism of the occurrence and development of sarcoma. [7] CMG complex is composed of the Cdc45 protein, four subunits of the GINS complex (Psf1, Psf2, Psf3, and Sld5) and the heterohexamer Mcm2-7 complex. It is most likely to be used as a replication helicase to unbind duplex DNA before the replication fork. [8]Recently, studies have shown that impaired function of the CMG complex will accelerate the development of genetic diseases. [9]The GINS complex consists of four subunits encoded by GINS1, GINS2, GINS3 and GINS4 genes. As a member of CMG, GINS complex plays important roles in the establishment of DNA replication forks and replisome progression.
We knew that GINSes were highly expressed in many tumors and were associated with their prognosis. For example, GINS1 was found to be a potential biomarker for patients with CRC, [10] and an inherited GINS1 de ciency was found to be one of the foundations of growth retardation [9]. GINS2 served as an independent risk factor for poor prognosis of early-stage cervical cancer through multivariate analysis. [11] Liu et al. found that GINS mRNA expression was signi cantly increased in patients with alcohol abuse. In experienments of mice, alcohol stimulated in ammatory cytokines and GINS2 knockdown further promoted them, while GINS2 knockdown promoted the alcohol-induced NF-κB activity of microglia, suggesting that GINS2 had a protective mechanism in alcohol-induced brain injury. [12] GINS3 has not only been linked to colon and lung cancer, [13,14] but also plays a central role in the eukaryotic replicative helicase CMG (Cdc45-MCM helicase-GINS). [15]GINS4 has also been associated with the progression and prognosis of lung and stomach cancer. [16,17] However, to our best knowledge, there is no speci c exploration about GINS with sarcoma. So, we tend to research on the expression and prognosis analyses of GINS in human sarcoma.

ONCOMINE Analysis
The online data-mining platform and cancer microarray database ONCOMINE (https://www.oncomine.org/) was used to analyze the transcription levels of GINS members among different cancers. [18] P value was generated by the Student t-test, the cut-off of p-value and fold-change were de ned as 0.05 and 2, respectively. and the mRNA expressions of the GINS family subunits in cancer samples was compared with normal ones. We also analyzed genes corrected with GINS members in sarcoma by using Oncomine. [19] GEPIA Analysis Gene Expression Pro ling Interactive Analysis (GEPIA) (http://gepia.cancer-pku.cn/) is a newly developed multi-function data analysis platform based on the Cancer Genome Atlas (TCGA) and the Genotype-Tissue Expression (GTEx). GEPIA includes 9736 tumors and 8587 normal samples. It was used to compare the different expression levels of GINS members between the human sarcomas and the normal samples. It also provided us survival analysis of sarcoma patients, including overall survival (OS) and disease-free survival (DFS), the patients were divided into two groups according to the median expression of GINS (high and low expression). We also calculated log-rank P-value and the cox proportional hazard ratio (HR) with 95% con dence intervals. [20] [21] Kaplan-Meier Plotter Kaplan-Meier Plotter (www.kmplot.com) is an online database which includes the expression levels of different genes and their prognostic values in pan-cancer. In our study, it was used to analyze the different relapse-free survival (RFS) of sarcoma patients with high GINS expression level and low GINS expression level. The hazard ratio (HR) with 95% con dence interval and the lop-rank P-value were also calculated. [22] CCLE Dataset Cancer Cell Line Encyclopedia (CCLE) (https://portals.broadinstitute.org/ccle) was established to deeply and accurately depict the genetic characteristics of cancer. It was between the Broad Institute and the Novartis Institutes for Biomedical Research and its Genomics Institute of the Novartis Research Foundation. CCLE conducted a large-scale in-depth sequencing of 947 human cancer cell lines covering more than 30 tissue sources, integrating genetic information such as DNA mutations, gene expression and chromosome copy numbers. The CCLE was an access to provide us the genomic data, visualization cell lines. The GINS family's expression in human sarcoma cell lines was obtained from the CCLE dataset. [23] Results

Transcriptional Levels of GINS members in Sarcoma Patients
Four members (GINS1, GINS2, GINS3, GINS4) of GINS family has been discovered in previous studies. They were found to be linked to many different types of tumor. The ONCOMINE database was used to compare the dissimilarity of transcription levels of GINS between cancer and normal samples. We detected that the transcription levels of GINS were upregulated in several cancers including cervical cancer, colorectal cancer, lung cancer and sarcoma. Especially, all the transcription levels of the four subunits of GINS were upregulated in sarcoma tissues ( gure 1). In table 1, as we showed, the mRNA expression level of GINS1 was signi cantly upregulated in sarcoma patients. In Detwiller Sarcoma dataset [24], the mRNA expression of GINS1 in pleomorphic liposarcoma was higher than in normal samples, with a fold change of 10.222. In brosarcoma, malignant brous histiocytoma and round cell liposarcoma, GINS1 was also more highly expressed than in normal samples, and the fold changes were 9.773, 10.455 and 6.478. The Barretina Sarcoma dataset [25] provided us that the expression levels of GINS1 were higher in pleomorphic liposarcoma (fold change = 4.135), myxo brosarcoma (fold change = 5.272), leiomyosarcoma (fold change = 4.566), dedifferentiated liposarcoma (fold change = 3.204) and myxoid/round cell liposarcoma (fold change = 3.412) than in normal samples.

Comparison of The Expression of GINS mRNAs in Sarcoma and Normal Tissues
Using the GEPIA dataset, we compared the expression of GINS mRNAs in sarcoma and normal tissues, it indicated that the expression levels of GINS1, GINS2 and GINS3 in sarcoma were statistically higher than in normal samples. The expression level of GINS4 in sarcoma was also higher than in normal samples, but with no signi cance. ( gures 2A-2F).

GINS Translational Factors Expression in Sarcoma Cell Lines
By interrogating CCLE, we detected the transcription level of GINS members in many types of cancer cell lines, and got the conclusion that GINS1, GINS2 GINS3 and GINS4 were highly expressed in sarcoma cell lines including Ewings sarcoma, osteosarcoma and chondrosarcoma. ( gures 3A-3D).

The Prognostic Value of GINS members in Sarcoma
The GEPIA and Kaplan-Meier Plotter was used to make the investigation of the prognostic ability of GINS1, GINS2, GINS3 and GINS4 expression levels in sarcoma. We can see from the ( gure 4A-4C), increased GINS1-3 were found to be linked to poor overall survival (OS) (p<0.05), disease-free survival (DFS) (p<0.05) and relapse-free survival (RFS) (p<0.05). Nevertheless, increased GINS4 was discovered tend to have signi cant association with poor RFS. To sum up, GINS1-4 were prognostic factors for sarcoma.

Discussion
The 21th century was actually the generation of molecular biology and biochemistry. Bioinformatics is a discipline using the methods of applied mathematics, informatics, statistics and computer science to study biology [27], mainly for biological mysteries of large and complex biological data [28]. Accordingly, we conducted our study using bioinformatics. The present study is the rst to use bioinformatics analysis to explore the expression and prognosis of GINS family in sarcoma. Furthermore, we probed into the correlation of all GINS family in sarcoma and its co-expressed genes. Dense-isotope substitution experiments discovered that GINS complex is important element rather than a passing traveler in initiation and process of replication [29]. Due to our research, a conclusion can be drawn that all the four subunits of GINS are overexpressed in sarcoma tissues, compared with normal tissues. GINS1, GINS2, GINS3 and GINS4 could make contributions to the treatment to sarcoma. Moreover, expression of all the four members encoding the subunits for GINS complex positively corrected with each other.
Due to previous study, Stable association of MCM complex and GINS complex form the CMG complex, which is responsible for the activity of DNA helicases. [30] The increased expression of GINS family members was involved in many different types of cancers and procedures [8][9][10][11][12][13][14][15][16][17]. It was associated with worse prognosis. Tang et al. discovered that the inhibition of GINS1 expression signi cantly weakened the proliferation in vitro, but accelerated the apoptosis of deviant cells [31]. Therefore, they de ned it as a target of osteosarcoma. We've described about that in mice experiment, knockdown of GINS1 expression caused inhibition of tumor growth [32]. On the contrary, overexpressed levels of GINS1 lead to high proliferation [33]. Through the research by Bu et al., the levels of GINS1 also in connected with the development of pancreatic cancer [34]. This is also con rmed by the research of Fu et al [35]. Our study indicated that the value of GINS1 in prognosis of human sarcoma was con rmed by 131 sarcoma patients. With the increased expression levels, there are also signi cant differences in DFS and OS.
GINS2 also found to be associated with tumors, by regulating PI3K/Akt and MEK/ERK signaling pathways, GINS2 facilitated cell proliferation, migration, invasion, and EMT in NSCLC [36]. Tian et al. also determined the identi cation of GINS2 as the hub gene though the WGCNA and PPI network. GINS2 was also reported to take part in DNA replication and cell progression as a companion of Sld ve 2 [37]. And another pathway has been found in non-small-cell lung cancer by Chi et al., it indicated that via p53/GADD45A pathway, GINS2 knockdown down-regulated cell proliferation, causes G2/M phase cell cycle arrest and augments apoptosis conversely. In human sarcoma, our results revealed that the expression level of GINS2 in sarcoma tissues was higher than it in normal tissues. By GEPIA dataset, we con rmed the prognosis value of GINS2, its increased expression was linked to poor DFS and OS.
As for GINS3, the expression level of Psf3, which is encoded by GIINS3, in the study of Tauchi et al. in 2016, was found to be association with age, gender, T factor, lymph node metastasis, stage and P factor in lung adenocarcinoma. Besides, poor DFS and OS were matched by higher expression of it [38]. After 3 years, they con rmed that Psf3 is the independent prognostic factor for lung adenocarcinoma through further discoveries by multivariate analysis. Additionally, they revealed that in the patients, who received postoperative chemotherapy, with high expression of Psf3 in stage 1, the ve-year survival rate was much greater. Conversely, the same result was not been observed in patients with low expression, which signi ed the Psf3 may serve as a biomarker of adjuvant tegafur-uracil administration to treat lung adenocarcinoma in stage 1 [39]. In another Japanese research, Psf3, as a member of CMG complex, accelerated excessive proliferation also in non-small-cell lung cancer [40]. These are consistent with our study, in human sarcoma, GINS3 may be a potential biomarker and its expression have in uence on the prognosis.
GINS4 was also reported expressed higher in gastric cancer, non-small-cell lung cancer, colorectal cancer and breast cancer than normal tissues [16,17,41,42]. There were different pathways, for example, in gastric cancer, via directly combination with Rac1/CDC42, GINS4 activated Rac1/CDC42 and then in uence the downstream pathways [16]. In colorectal cancer, the KLF/GINS4 might be the crucial factor [17]. Our study found high expression of GINS4 in human sarcoma tissues, and it tended to have tight connection with OS and DFS, but failed to achieve signi cance.
The e ciency and convenience of bioinformatics are obvious, the datasets we used helped us to unearth the expression and prognosis of GINS family in human sarcoma. There are also limitations of our study, we used almost every practicable data, it still seems to be inadequate, more future studies and more datasets are needed. We didn't nd the pathways of GINS family in human sarcoma, and we will keep on and achieve Immunohistochemical experiments.

Conclusions
We can come to the conclusion, through our study, that high expression of GINS members in sarcoma was closely related to sarcoma tumorigenesis. The prognostic relevance tested was also obvious, increased expression of GINS1, GINS2, GINS3 and GINS4 may provide us new therapeutic targets for the treatment of sarcoma. All the GINS members could be seen as potential prognostic markers to improve prognostic accuracy of sarcoma.

Declarations
Ethics approval and consent to participate Not applicable.

Consent for publication
Not applicable

Competing interests
The authors declare that the research was conducted in the absence of any commercial or nancial relationships that could be construed as a potential con ict of interest.

Authors' Contributions
Gen Wu and Wanchun Wang conceived, designed and conducted the experiments. Gen Wu and Ziyuan Chen wrote the paper. Gen Wu and Ziyuan Chen collected the data. Gen Wu and Wanchun Wang edited the paper. Gen Wu and Ziyuan Chen provided the research guide. Wanchun Wang supervised this project. All authors read and approved the nal manuscript.