The expression and prognostic value of minichromosome maintenance markers in human breast cancer: a comprehensive analysis

Objective : Breast cancer (BC) is one of the most health-threatening neoplasms for women worldwide. Despite advances in detection and treatment strategies over the past few decades, the current biomarkers of BC are less than satisfactory for effective prognosis and individualized treatment. This study aimed to investigate the new biomarkers to meet this urgent demand. Methods : The current study investigated the transcriptional levels of minichromosome maintenance genes (MCMs) in BC patients from the Oncomine, UALCAN database, and Gene Expression Profiling Interactive Analysis (GEPIA); protein expression levels of MCM proteins in BC patients were derived from the Human Protein Atlas (HPA) database. Further, survival analysis was evaluated with Kaplan-Meier Plotter. BC genome atlas data were obtained from cBioPortal databases. Gene regulatory network analysis was performed using the STRING online tool, and gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses were performed using DAVID. Results : Based on multiple database analysis, mRNA and protein levels of MCM2 , MCM4 and MCM10 were much higher in BC patient, and survival analysis showed that high transcription levels of most MCMs were found to be associated with poor prognosis for BC patients; moreover, the MCMs genetic alterations, especially of MCM2 , MCM4 and MCM10 , were found in 45% of BC patients. In addition, dysregulation of MCMs was considered to possibly affect DNA damage/repair, cell cycle dysregulation and chromosome instability. Conclusions : In summary, this study indicated that MCM2 , MCM4 , and MCM10 are potential prognostic markers and therapeutic targets for BC.


Introduction
Breast cancer (BC) is the most common type of malignancy and the second leading cause of cancer-related mortality in women worldwide [1,2]. Despite advances in detection and treatment strategies over the past few decades, the prognosis for BC patients is still suboptimal, with only one-fifth surviving for 5 years [3][4][5][6]. The current biomarkers of BC are less than satisfactory for effective prognosis and individualized treatment; thus, new biomarkers for BC are urgently needed.
DNA replication is involved in biological processes such as development, aging and cancer etiology [7]. The minichromosome maintenance gene (MCM) family members are involved in cell cycle processes and DNA replication [8]. This gene family comprises 10 members: serum response factor (SRF, as know as MCM1), MCM2-MCM10 [9]. Assembled by six subunits (protein MCM2-MCM7), the MCM complex exhibits helicase activity and participates in DNA replication initiation [10][11][12][13]. Although MCM1, MCM8, MCM9 and MCM10 protein are not included in the MCM complex, they are also indispensable for DNA replication [14,15].
Mounting evidence has shown that MCM family members are dysregulated in various malignant tumors, and can be used predict tumor progression and prognosis [16][17][18][19][20][21][22][23]. High mRNA expressions of MCM2, MCM3 and MCM7 are of great relevance as they indicate poor prognosis for patients with glioma [20]. MCM6 protein has been identified as a driver of S/G2 cell cycle progression; thus, the upregulation of MCM6 transcriptional level is indicative of adverse tumor features and poor outcomes in hepatocellular carcinoma [21]. The upregulation of MCM10 transcriptional level is strongly linked to poor overall survival (OS) for lung cancer [22]. Further, the upregulation of MCM2, MCM4, MCM6 and MCM10 in pancreatic cancer correlate with poor prognosis [23]. MCMs play complex and distinct roles in human BC [24,25]. However, the clinical significance and the particular functions of MCMs in BC have not yet been fully elucidated, and required further exploration. Therefore, we performed our study to analyze the expression levels and genetic alterations of MCMs in BC patients in detail, to determine the expression patterns, potential functions and prognostic values of these markers.

MCM-expression analysis
The Oncomine database (http://www.oncomine.org) was used to analyze the mRNA expression levels of MCMs. The comparison of mRNA expression levels of MCMs between BC and normal samples was conducted using the Student's t-test. The cut-off values for the p-value and foldchanges were as follows: p-value < 0.01, fold changes >2. GEPIA (http://gepia.cancer-pku.cn/) was used to validate the transcription levels of MCMs in BC tissues and different pathological stages, as well as to perform the correlation analysis of MCMs with each other. The UALCAN database (http://ualcan.path.uab.edu/) was used to detect the transcriptional expression levels of MCMs in different BC subtypes.

Kaplan-Meier plotter analysis
The Kaplan-Meier plotter (http://kmplot.com/analysi s/) was used to evaluate the correlation between MCM expression and prognosis in BC. BC patients were divided into high and low expression groups according to the median value, and the OS, relapse-free survival (RFS) and distant metastasis-free survival (DMFS) of BC patients were assessed using Kaplan-Meier survival plots. The hazard ratios (HRs), 95% confidence intervals (CIs) and log rank p-values less than 0.05, were retrieved from the Kaplan-Meier plotter.

Immunohistochemistry analysis
The Human Protein Atlas (HPA) database (https://ww w.proteinatlas.org/) is a valuable tool for studying protein expression profiles in normal tissues, pathological tissues and cell lines [26]. MCM protein expression of BC and normal tissues was determined from the HPA database.

Mutation and copy-number alteration analysis
The Breast Invasive Carcinoma (TCGA, Firehose Legacy) database, involving 1101 cases, was selected for further analyses of MCMs using cBioPortal (http://www.cb ioportal.org). Mutations, putative copy number alterations (CNAs) from the genomic identification of significant targets in cancer (GISTIC), mRNA expression Z scores relative to diploid samples (RNA-seq v.2 RSEM) and protein expression Z scores (reverse phase protein array [RPPA]) were included in the genomic profiles.

Gene regulatory network analysis
To explore the interaction among MCMs at the protein levels, the STRING online tool (https://string-db.org/) was used to construct a protein-protein interaction (PPI) network with a confidence score of 0.4. DAVID (https: //david.ncifcrf.gov/) was used for gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses of MCMs. The related information of their involvement in biological processes (BPs), molecular functions (MFs), cell components (CCs) and different pathways was identified based on p-values < 0.05.

Transcriptional levels of MCMs in BC
The Oncomine database contains and compares available microarray data from multiple cancer types, including BC, with those from normal samples. The expression of MCMs in cancer cells is shown in Fig. 1 [27][28][29][30][31], followed by MCM2 in 8 datasets [28,29,31] and MCM10 in 7 datasets [28,29,31]. However, owing to the relatively small sample size, whether the expression of MCM5, MCM6 and MCM8 was elevated in BC remains controversial.

Relationship between MCM mRNA and protein expression levels in BC
Further investigation of the transcriptional levels of the MCMs in BC and in different subtypes using GEPIA online tool and UALCAN database showed that MCM2, MCM4 and MCM10 were significantly upregulated in the tumor tissues compared with expression in normal tissues, but there was no significant difference in the expression of the other MCMs between BC and normal tissues ( Fig. 2A); when divided into different subtypes of breast invasive carcinoma, the results showed that almost all MCM family members were significantly upregulated in these subtypes compared with levels in normal sample, except MCM1 and MCM9, which were downregulated; of these, the expression levels of MCM1 in triple negetive BC and MCM9 in luminal and triple nagetive BC were not significantly different compared with those in normal tissues (Fig. 2B). IHC (Immunohistonchemistry, IHC) staining images for MCM proteins in BC and normal sample obtained from the HPA database demonstrated that MCM2, MCM4 and MCM10 protein levels were more highly expressed in BC tissue than in their counterparts, whereas others were not significantly different in BC tissue (Fig. 3).

Association between mRNA expression of MCMs and BC patient prognosis
To further investigate the critical role of MCMs in predicting the prognosis of BC patients, we analyzed the association between MCMs transcription levels and BC pa-tients survival using Kaplan-Meier plotter analysis. The BC patients were divided into high and low expression groups according to the median value. High expression levels of MCM family members were significantly correlated with OS, RFS and DMFS: MCM2, MCM4, MCM5, MCM6, MCM7 and MCM10 correlated with worse OS, RFS, and DMFS, MCM8 correlated with the poor DMFS, MCM9 correlated with better OS and RFS, and MCM1 correlated with better RFS (Fig. 4, Supplementary Fig. 1). In addition, we found that the expression levels of MCM2, MCM3, MCM7 and MCM10 varied significantly in different tumor stages of BC, indicating that they are associated with clinical stages (Fig. 5).

Alteration to MCMs in BC and correlations among them
We next analyzed MCM alterations using the cBio-Portal, which is an online database for Cancer Genomics, providing images and analyses of large-scale cancer genomics datasets [32]. As shown in Fig. 6, the most common genetic change among 1101 BC patients was high mRNA expression (Fig. 6A); MCMs were altered in 494 BC patients (45%) and multiple alterations were detected in 129 (11.8%) (Fig. 6B). Among them, MCM4, MCM10 and MCM2 alterations were associated with 23%, 16%, and 11% of BC cases, respectively surpassing the alterations to other MCMs. The relationship between MCM mRNA levels was analyzed using the GEPIA online tool coupled with Pearson's correction (Fig. 6C). In BC, a close correlation between the mRNA expression of MCM2-MCM7, but not between MCM1 and the other nine MCMs was found, and the following MCMs showed significant and positive correlations: MCM8 with MCM2, MCM3, MCM5, MCM6 and MCM10; MCM9 with MCM4; MCM10 with MCM2-7 and MCM8. Among them, MCM2 and MCM6 exhibited the highest positive correlation with a Spearman's correlation coefficient of 0.77.

MCM-regulated biomolecular network analysis of BC genomics
To explore how MCMs exert their regulatory effects on the occurrence and development of BC, potential MCMregulated genes were obtained from the STRING online tool. MCM-related gene network showed that MCMs had a close correlation with 50 genes, which were cell cycle-related or involved in DNA damage/repair, including POLA1, ORC6, RPA3, CDC7 and RFC2 (Fig. 6D). Further functions related to pathways of MCMs and their frequently altered neighbor genes were predicted by analyzing GO and KEGG processes in the DAVID. GO enrichment analysis is based on three aspects to predict the functional roles of target host genes. These results indicated that the MCMs in the BPs of GO enrichment were markedly related to DNA replication, telomere maintenance, telomere organization, and cell cycle transition (Fig. 7A). For CCs of GO en-  richment, the MCMs were particularly enriched in nuclear chromosome parts, chromosomal regions, replication forks and protein-DNA complexes (Fig. 7B). In addition to MF, MCMs were remarkably associated with chromatin binding, damaged DNA binding, single-stranded DNA binding and telomeric DNA binding (Fig. 7C). Moreover, the results of KEGG analysis demonstrated that MCMs were enriched in the cell cycle, DNA replication, nucleotide excision repair, mismatch repair, and homologous recombination (Fig. 7D).

Discussion
The proliferation of neoplastic cells is usually accompanied by dysregulated DNA replication [33]. As an essential molecule in DNA replication, MCM protein were found to be a determinant of the initiation and progression of malignancy [34]. MCM mutations appear more frequently in neoplastic cells than in normal cells [35], which leads to chromosome loss, DNA damage and increased recombination [36,37]. The role of MCM factor dysregulation in the tumorigenesis and prognosis of cancers has been partially confirmed, and further bioinformatics analysis of BC had not yet been performed.

MCM1
, as a member of the MADS box transcription factor family, affects processes such as cell cycle, growth, differentiation and apoptosis by regulating the activation of neighbor genes [38]. The current study revealed that the downregulation of MCM1 was significantly associated with HER2 positive BC and luminal BC, and with poor RFS.
The MCM 2-7 complex is a major factor in DNA replication initiation and elongation [39]. Our results showed that the transcriptional levels of all members of MCM complex in BC and its subclasses were upregulated and that they were closely correlated with each other, which could indicate their regulation based on similar transcription factors or signaling pathways. Combining the analysis of multiple databases, the mRNA expression levels of MCM2, MCM4 and MCM10 were significantly upregulated, whereas others were only slightly increased in BC tissue. Upregulated mRNA expression of MCM2, MCM4 and MCM10 was in accordance with protein expression levels and was associated with poor OS, RFS and DMFS. The genetic alteration rates of MCM4, MCM10 and MCM2 were highest, which might be the reason for their mRNA upregulation. As a member of the MCM protein complex family, the mRNA expression of MCM2 was firmly established as instrumental in DNA replication and cell proliferation [40,41]; therefore, the precise control of MCM2 expression is crucial for the maintenance of genomic stability [42]. It has been reported that the knockdown of MCM2 prevents tumor cells from proliferating [43], whereas high expression of MCM2 is positively associated with tumor size [44]. The present study also found that MCM2 was correlated with tumor stage, which is in agreement with the findings of Wojnar et al. [24]. The MCM4 gene mutation destabilizes the MCM 2-7 complex, causing DNA replication impairment and chromosome instability, which indicates its causative role in BC progression [45,46]. A previous study showed that MCM4, for which a mutation was determined to be a causative factor in BC progression, could be a potential novel prognostic and predictive indicator [19,46]. Although not included in MCM 2-7 complex, MCM10 has a critical role in the initiation of DNA replication and tumor progression through interactions with the MCM 2-7 complex [47,48]. MCM10 was reported to induce BC metastasis via the Wnt/β-catenin pathway and can be defined as a potential diagnostic biomarker and a promising target for BC [25]. However, Kwok et al. [19] demonstrated that the converging overexpression of more than four MCMs at the mRNA level is associated with a significantly higher risk of poor prognosis.
As to other members of the MCM 2-7 complex, the expression levels of MCM3, MCM5, MCM6 and MCM7 differed significantly only between normal tissue and each major subclass of invasive BC. MCM3 protein was reported to be a contributing factor to cell proliferation, and high MCM3 expression is association with poor prognosis in invasive ductal carcinoma patients [49]. However, no association between MCM3 expression and BC patient outcomes was found in the present study. However, a significant correlation was observed between MCM3 mRNA expression and tumor stage. MCM5 is involved in the cell cycle regulation [50]. High expression level of MCM5 was proved to be significantly associated with poor prognosis in BC patients in the current study. MCM6 also appears to have a prognostic value. In our report, MCM6 upregulation was found to be significantly associated with poor prognosis in BC patients, which is in agreement with the study by Kwok, H. F. et al. [19]. As a licensing factor crucial for DNA replica-tion, when MCM7 protein is indirectly phosphorylated by activated epidermal growth factor receptor, its phosphorylation increases its association with other MCM proteins, thereby promoting DNA synthesis complex assembly, together with cell proliferation, which is correlated with poor BC patient survival [51]. In the current study, high expression level of MCM7 was associated with tumor stage and adverse prognosis.
Of the remaining members of the MCM family, MCM8 might be associated with chromosomal instability [52]. The current study showed that high MCM8 expression is associated with poor prognosis. Little is known about the role of MCM9 in BC. The prognostic value of MCM9 for BC has not been investigated yet; however, its upregulated was found to be associated with better OS and RFS in our study.

Conclusions
In conclusion, we analyzed the mRNA and protein expression, gene alterations, prognostic values, and functions of MCMs in BC. Our results indicated that the mRNA and protein expression of MCM2, MCM4 and MCM10 was upregulated in BC patients, which was associated with poor OS, RFS and DMFS. Furthermore, genetic alterations were comparatively more frequent for MCM2, MCM4, and MCM10, which themselves appeared to directly affect DNA replication impairment and chromosome instability in BC, or indirectly affect these processes-through mutations affecting adjacent cell cycle-related genes or those involved in DNA damage/repair. Our findings suggest that MCM2, MCM4 and MCM10 are potential prognostic markers and therapeutic targets for BC.

Author contributions
JX-Performing the research, visualization, writing-Original draft preparation. YZ-Designing the study, data analysis, writing-Original draft-review & editing. YG-Designing the study, conceptualization, Data curation.

Ethics approval and consent to participate
Not applicable.