Identification of the DNA Replication Regulator MCM Complex Expression and Prognostic Significance in Hepatic Carcinoma

Background The microliposome maintenance (MCM) complex, MCM2-7, is revealed to be involved in multiple cellular processes and plays a key role in the development and progression of human cancers. However, the MCM complex remains poorly elaborated in hepatic carcinoma (HCC). Methods In the study, we found the mRNA and protein level by bioinformatics. We also explored the prognostic value, genetic alteration, interaction network, and functional enrichment of MCM2-7. The MCM expression and correlation among these MCMs in HCC cell lines were identified by western blot. Results MCM2-7 was significantly increased in HCC tissues compared to normal liver tissues. The high level of MCM2-7 had a positive correlation with poor prognosis. However, MCM2-7 alterations were not correlated with poor OS. MCMs were both increased in HCC cell lines compared to the normal hepatocyte cell line. Furthermore, the positive correlation was found among MCMs in HCC cell lines. Conclusions The MCM complex was increased in HCC tissues and cell lines and negatively correlated with prognosis, which might be important biomarkers for HCC.


Introduction
Hepatic carcinoma (HCC), a severe malignant disease of the digestive system, ranks sixth in terms of morbidity (over 0.8 million new cases) and fourth in mortality overall (over 0.7 million deaths) in 2018 [1]. There are two important risk factors for HCC, including alcohol consumption [2] and hepatitis virus [3]. Although the development of new targeted drugs and multidrug combinations has improved the battlefield for HCC, only about 5-14 percent of patients with HCC have a five-year survival rate [4,5], with the development of bioinformatics and high-throughput sequencing technology, inhibitor of apoptosis protein (IAP) family members [6], kinesin family members [6], cytochrome P2C (CYP2C) subfamily members [7], CDK1, PBK, RRM2, and ASPM [8]. Recently, Han et al. found through bioinformatics that SCAMP3 may be an important marker in the development of liver cancer [9]. Therefore, it is very important to explore the correlation between prognostic value and new gene complexes in HCC.
The microliposome maintenance (MCM) protein was first discovered in sacs cerevisiae, and the mutants showed defects in microliposome maintenance, which play a key role in DNA replication [10]. MCM2-7 is a group of six structurally related proteins, from yeast to humans, which are highly conserved and interact to form a hexamer. In the progression of DNA synthesis, MCM2-7, as nuclear proteins, binds to chromatin by a cell cycle specific manner, resulting in promoting cell proliferation and helicase activity [10]. Disorders in these proteins can directly disrupt the DNA replication system, leading to cancer occurrence, development, and progression [11].
The MCM complex acts as an important regulator in multiple pathophysiological processes, including DNA replication [12], cell cycle [11], proliferation [13], migration [14], invasion [14], immune response [15], and apoptosis [11]. It has been previously reported in the literature that MCM is highly expressed in many cancers compared to normal tissues. For example, MCM2 got a high level in HCC, oral squamous cell carcinoma [16], gastric cancer [17], breast cancer [18], colon cancer [19], and ovarian cancer [20]. MCM2 was considered as a potential therapeutic target for cancer treatment, and the level of MCM2 could predict poor prognosis for osteosarcoma [21], gastric cancer [22], lung adenocarcinoma [23], diffuse large B cell lymphoma [24], and esophageal cancer [25]. Recent research suggested that MCM2 might be a potential therapeutic target for HCC [26]. Furthermore, Deng et al. found that MCM2 inhibition could increase the sensitivity of carboplatin in ovarian cancer cell [27]. MCM3 had similarly a high expression level in multiple cancer types, such as osteosarcoma [21], salivary gland tumors [28], and glioma [29]. Ha et al. indicated that MCM3 got a high expression in leukemia, lymphoma, uterine cervix cancer, colon cancer, lung cancer, gastric cancer, kidney cancer, breast cancer, and malignant melanoma [30]. MCM4 levels were elevated in esophageal cancer [31], uterine cervical carcinoma [32], and non-small-cell lung cancer [33]. MCM5, as another regulator in DNA replication, was overexpressed in colon cancer [19], oral squamous cell carcinoma [34], cervical cancer [35], thyroid cancer [36], and bladder cancer [37]. The expression of MCM6 was found to be enhanced and its high level had a close relationship with unfavorable prognosis in colorectal cancer [38], breast cancer [39], AO [15], HCC [14], endometrioid adenocarcinoma [40], lung cancer [41], meningiomas [42], cervical cancer [43], Hodgkin's lymphoma [44], and Merkel cell carcinoma [45]. Ectopic expression of MCM7 has been indicated that can promote the progression of prostate cancer [46], HCC [47], breast cancer [48], and acute myeloid leukemia [49]. These studies sufficiently indicated differential level of MCMs in multiple cancer types, but few studies systematically focused on the prognostic value of the whole MCM complex members in carcinogenesis. In this study, we comprehensively demonstrated the transcriptional level of MCMs and found its prognostic value in HCC. Moreover, we also analyzed the interaction network, genetic mutation, and functional enrichment of MCMs by bioinformatics. , an open online database of protein expression profiles, assists researchers in studying the differential expression of proteins between cancer and normal tissues.

Clinical Samples.
A total of 30 HCC tissues were surgically resected in the Affiliated Nanhua Hospital, University of South China (Hengyang, Hunan, China), from 2010 to 2014. These tissues were made to 3 pieces of 10 × 10 chips. The collection and use of tissues followed the procedures according to the ethical standards as formulated in the Helsinki Declaration. And written informed consent was obtained from each patient, which was approved by the research ethics committee of University of South China. All patients did not receive radiotherapy or chemotherapy.
2.5. Kaplan-Meier Plotter Analysis. KM plotter (http:// kmplot.com/analysis/), a survival database of patients, can analyze survival curves for many types of cancer [6]. The KM plotter is utilized to assess the prognostic value of MCMs in HCC.
2.6. GeneMANIA and STRING Analysis. GeneMANIA (http://genemania.org) [52] and STRING (https://string-db .org/) [53] are web tools to identify the interactions between genes/proteins, respectively. The interactions between MCMs and other gene/proteins are utilized by GeneMANIA and STRING at the gene or protein level. 2.7. cBioPortal for Cancer Genomics Analysis. cBioPortal for Cancer Genomics (http://www.cbioportal.org/), an open and free web tool, can be used to interactively explore multiple cancer genome datasets [54]. The correlation between MCM alterations and survival outcome in patients with HCC was analyzed by cBioPortal.
2.8. Metascape Analysis. Metascape (http://metascape.org) is an online analysis website available at enrichment pathway analysis and gene function annotation, which can be used to analyze the pathway and process enrichment of MCMs and the 40 neighboring genes [ 2.11. Immunohistochemistry. According to the manufacturer's instructions (Maixin Biotech. Co., Fuzhou, China), the slides were incubated with the primary antibody (diluted 1 : 100) at 4°C overnight, and normal rabbit immunoglobulin G was the negative control. The score of positive staining degree and percentage of stained cells were as follows: 0, no staining; 1, light brown; 2, dark brown and 0, stained cells < 5%; 1, stained cells range from 5% to 25%; 2, stained cells ranged from 26% to 50%; 3, stained cells > 50%. Scores were obtained by increasing the strength and reactivity of the reaction. A score of 2 is defined as high expression, and a score below 2 is defined as low expression.
2.13. Statistical Analysis. Statistical analyses were performed in the R Programming Language (version 3.6). All statistical tests were bilateral, and P < 0:05 was statistically significant.

The mRNA and Protein Expression of MCMs in HCC.
Firstly, we utilized the Oncomine database to extract the data of MCM transcriptional levels in different cancer and corresponding normal tissues ( Figure 1). MCM4 was also enhanced in HCC compared to normal tissues based on Wurmbach Liver datasets [3]. Chen Liver datasets showed an obviously increased MCM6 level in HCC [57]. In addition, Roessler Liver and Roessler Liver 2 datasets indicated that MCMs were both significantly increased in HCC compared to normal tissues [58]. The statistical significance results with corresponding P values are shown in Figure 1 and Table 1.
We also used GEPIA to compare the transcriptional levels of MCMs in HCC and normal tissue ( Figure 2). We found that the expression of both MCM proteins in tumor tissues was significantly upregulated. Moreover, the correlation between MCM level and HCC stages was also analyzed in GEPIA, which indicated that both MCMs were closely associated with HCC stage (Supplementary Figure S1).
In addition, the immunohistochemistry (IHC) staining images for MCM protein in HCC and normal liver tissues were extracted from the HPA database ( Figure 3). We found that these proteins were both significantly increased in HCC tissues compared to normal liver tissues.
The further analysis of these impacts on prognosis by these proteins is provided. We found that the prognosis significance of MCMs had a close correlation with some clinicopathological parameters, including clinical stages, pathology grade, and vascular invasion (Table 3). High expression of MCM2 was prominently associated with worse OS in HCC stage 1+2 (HR = 2:13, P = 0:0019) and stage 3+4 (HR = 2:09, P = 0:043). Likewise, similar results on stage 1+2 and stage 3 +4 were also observed in MCM3 (HR Both OS significance of MCMs had a significant correlation with no vascular invasion, which indicated that high 3 BioMed Research International level of MCMs could predict the poor prognosis in HCC patients without vascular invasion. In the HCC patients with vascular invasion, only MCM7 could suggest a poor prognosis (HR = 2:27, P = 0:0328). Furthermore, the OS significance of MCM2 was not associated with grade 1 (HR = 2:43, P = 0:0666), but was associated with grade 2 (HR = 1:82, P = 0:0368) and grade 3 (HR = 4:13, P = 0:000088). Other MCMs are well summarized in Table 3. Next, we analyzed the association between prognosis significance of MCMs and hepatitis virus infection and alcohol consumption, respectively. The results indicated that both MCM overexpression had a positive association with poor OS in HCC patients without hepatitis virus infection or alcohol consumption. Only MCM5 and MCM7 had a significant correlation with poor OS in HCC patients with alcohol consumption. MCM6 could predict the poor OS (HR = 2:04, P = 0:0316), RFS (HR = 1:74, P = 0:0258), PFS (HR = 1:7, P = 0:0218), and DSS (HR = 2:26, P = 0:0469) in HCC patients with hepatitis virus infection. The detailed results are summarized in Table 4.

Correlation Analyses of MCMs in HCC Patients
. We extracted the MCM mRNA level between each other in HCC from TCGA Provisional dataset (RNA Seq V2 RSEM) by using the cBioPortal. Spearman's correlation analysis among these MCM levels suggested significantly positive correlation between both MCM2/3/4/5/6/7 and other MCMs ( Figure 5(a)).   Figure 1: Oncomine analysis of MCMs in different cancer types. MCM expression between different cancer types and corresponding normal tissues was summarized. Threshold (P value ≤ 0.05; |FDR | ≥2; gene rank ≤ 10%; data type: mRNA) is expressed in colored cells. In tumor tissue, red cells represent overexpression of the target gene compared to normal tissue, while blue cells represent downregulation of the gene. Gene levels are indicated by the color depth of the cell. (You can refer to the web version of this article to explain the color reference in the diagram.) 4 BioMed Research International

BioMed Research International
Next, we utilized the GeneMANIA tools to analyze the association of MCMs at gene level ( Figure 5(b)). This result indicated that the physical interactions among MCM2-7 were significant in this network, which might attribute to the shared protein domains. Relationships were significantly found among MCM2-7 in coexpression. Furthermore, pathway was noticed in reactome among MCM2-7 and other key genes, including CDC45, CDC7, ORC6, MCM10, CDT1, ORC5, GINS4, CLSPN, ORC4, and POLD3.
We further identified the protein interactions of MCM2-7 by the STRING database ( Figure 5(c)). The interactions among MCM2-7 were shown in experiments, databases, and coexpression. Moreover, the network for MCM2-7 and the 40 altered neighboring genes was constructed, such as

Functional Enrichment Analysis of MCMs in HCC.
Finally, we excavated GO and KEGG pathway data for MCMs and their 40 altered neighboring genes by using Metascape. Top 5 KEGG pathways were DNA replication, cell cycle, homologous recombination, pyrimidine metabolism, and viral carcinogenesis (Figures 6(a) and 6(b)  Biological processes are as follows: DNA replication, DNAdependent DNA replication, DNA replication initiation, nuclear DNA replication, double-strand break repair via homologous recombination, DNA strand elongation involved in DNA replication, telomere maintenance via semiconservative replication, regulation of DNA replication, DNA replication checkpoint, DNA replication preinitiation complex assembly, negative regulation of DNA replication, protein localization to chromosome, chromosome separation, G2 DNA damage checkpoint, and histone phosphorylation; cellular components are as follows: replication fork, chromosome, telomeric region, replication fork protection complex, chromatin, and centrosome. Next, the proteinprotein interaction enrichment analysis revealed that biological functions were mostly connected with the activation of ATR in response to replication stress, activation of the prereplicative complex, DNA replication preinitiation, processing of DNA double-strand break ends, HDR through homologous recombination (HRR) or single-strand annealing (SSA), and homology directed repair (Figures 6(e) and 6(f)).  Figure S2). The next section of the survey was concerned with MCM expression in HCC tissues. The results, as shown in Figure 7(c), indicated that MCMs were significantly enhanced in HCC tissues compared to paracarcinoma tissues. Together, these results provided important insights into the fact that MCMs might play significant roles in the formation, development, and progression of HCC.

Discussion
Currently, more and more studies indicated that ectopic expression of MCMs could promote DNA replication [12] and accelerate cell cycle [11] and metastasis [14]. MCMs were involved in the development and progression of many  [59]. In previous studies, the heterohexameric complex composed of MCM2-7 has been well summarized in human cancer cells. However, there are still many questions to be systematically solved about the expression, function, interaction, and prognostic value of MCMs in HCC. Therefore, we conducted a comprehensive analysis to reveal the transcriptional level, function enrichment, gene/protein interaction, and prognostic values of MCM2-7 in HCC.
In this study, we found that the expression of MCM2-7 was significantly increased in HCC tissues compared to normal liver tissues. MCM2, an oncogene, was correlated with the development and progression from cirrhosis to HCC [60]. MCM2 protein strongly expressed in high-grade squamous intraepithelial lesion may be useful as a cascade screening tool for detecting precancerous changes in cervical cancer [61]. Our results indicated that high level of MCM2 was significantly associated with worse OS/RFS/PFS/DSS in HCC patients without hepatitis virus infection. Zhao et al. revealed that MCM3 was a better marker of proliferation than Ki67, making it a valuable prognostic tool independent of ER and HER2 states [62]. In the group of nonalcohol consumption, our study indicated that HCC patients with elevated MCM3 expression had a bad OS/RFS/PFS/DSS. Choy et al. indicated that MCM4 could be used as a more sensitive proliferative marker for the identification of esophageal lesions [63]. We found that HCC patients with elevated MCM4 mRNA levels had unfavorable RFS and OS. Gong and his colleagues revealed that ectopic expression of MCM5 had a close correlation with malignancy and poor prognosis, which might be a potential prognostic marker in renal cell carcinoma [64]. In the study, we found that high level of MCM5 had also close correlation with the poor prognosis of HCC, especially in OS/RFS/PFS/DSS. Liu et al. suggested that MCM6 could indicate poor prognosis and promote migration and invasion, which could be predicted preclinical early recurrence in HCC patients to indicate more careful monitoring and aggressive treatment intervention [14]. Similarly, we also found MCM6 expression predicted poor OS and PFS in stage 1+2, stage 3+4, grade 1/2/3, or nonvascular invasion patients. Furthermore, MCM7 has advantages over traditional cell cycle markers, such as Ki67 and PCNA, because it has a higher sensitivity and is less susceptible to external factors, including inflammatory factors [65]. Given that Ki67 and PCNA expression can only be observed at certain stages of replication and can be easily interfered with, the presence of MCMs at all cellular stages may be the reason for the advantage [66]. Likewise, we analyzed the correlation between MCM7 level and prognosis, which indicated that MCM7could also be a valuable prognostic marker for HCC patients.
The relationship between MCMs and genetic alteration was observed in HCC using the cBioPortal database. However, the HCC prognosis was independent of the mutation, which might indicate that ectopic expression of MCMs was induced by other ways in HCC, such as protein phosphorylation, slunoylation, and ubiquitination. The epigenetic modifications, such as DNA methylation, histone acetylation, and noncoding RNA regulation, were still unclear in the MCM complex. HCC, as an acquired disease, may be more due to the epigenetic modifications and abnormal molecular signal transduction than to genetic factor and gene alteration [4,5]. Therefore, further analyses of protein and epigenetic modification are needed for the comparison with the analyses of gene alteration. The mechanism of gene alteration of these MCMs also needed further exploration.
In order to further explore the related functions and signaling pathways of these proteins, we studied the functional enrichment of MCMs and its mechanism by Metascape. The results indicated that the pathways involved in MCMs might contain DNA replication, cell cycle, homologous recombination, pyrimidine metabolism, and viral 9 BioMed Research International carcinogenesis. These pathways were frequent disorder in carcinogenesis. For example, Lin and his colleagues found that DNA replication could accelerate the cell cycle to promote carcinogenesis by the MCM complex [68]. Breast cancer type 1 susceptibility protein (BRCA1), a tumor suppressor, induces DNA double-strand break repair by homologous recombination, protecting DNA replication forks from attrition [69]. Sweeney et al. found that the combination of glutamine and glutamine-derived metabolites in purine and pyrimidine synthesis was inhibited by dimethylaminopurine and rhodoxin, effectively blocking the key biosynthetic pathway for the survival of leukemia cells [70]. Viral carcinogenesis has been demonstrated in nasopharyngeal carcinoma [71] and HCC [72] by EBV and hepatitis virus infection,

12
BioMed Research International respectively. However, more work and experiments are needed to verify these bioinformatics predictions, which will help to investigate the role of MCM2-7 and related signaling pathways in the development of HCC.

Conclusion
In this study, we systematically summarized the mRNA and protein level of MCMs and useful prognostic information about MCMs in HCC. Furthermore, we also analyzed the genetic alteration, coexpression, gene/protein network, and GO/KEGG enrichment analysis of MCMs. Relevant results indicated that the mRNA and protein level of MCMs was significantly increased in HCC tissue compared to normal liver tissues. KM plotter analysis showed that high expression of MCMs indicated a worse OS/RFS/PFS/DSS in HCC patients. Importantly, HCC patients with MCM alteration did not display worse OS compared with the ones without MCM  14 BioMed Research International alteration, which might be attributed to ectopic expression of MCMs mediated by other molecular mechanisms. In conclusion, MCMs could be an effective prognostic marker for HCC. Our results can help to better understand the pathogenesis of HCC and develop more effective clinical treatments in the future.

Data Availability
The data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest
The authors declare that they have no conflicts of interest.

Authors' Contributions
The conception and design were devised by Ting Cao, Shi-jie Yi, and Li-xin Wang

Supplementary Materials
Supplemental Figure S1: the expression of MCMs in different clinical stages. Supplemental Figure S2: