Minichromosome maintenance gene family: potential therapeutic targets and prognostic biomarkers for lung squamous cell carcinoma

Aberrant expression of MCM family members in LUSC

The Tumor Immune Estimation Resource (TIMER) database was used to examine transcriptional levels of MCMs and to compare their expression in diverse cancer types to those in healthy tissues. The mRNA expression of MCM2–10 was substantially higher in LUSC tissues than in healthy tissues, whereas MCM1 levels showed no significant difference (Figure 1A–1J). Gene Expression Profiling Interactive Analysis (GEPIA2), which contains resources distinct from those in the TIMER database, was used to further explore the mRNA expression of the MCM family genes. We found that all MCM factors, except for MCM1 and MCM9, exhibited considerably higher expression in LUSC than in normal tissues (P < 0.05) (Figure 1K).

Figure 1. Expression of minichromosome maintenance (MCM) family members in lung squamous cell carcinoma (LUSC). (A–D) Expression of MCM1–10 in pan-cancer. *p < 0.05, **p < 0.01, ***p < 0.001 compared with control.
(E–J) Expression of MCM1–10 in pan-cancer. (K) Expression of MCMs in LUSC. *p < 0.05, **p < 0.01, ***p < 0.001 compared with control.

We further investigated the results of immunohistochemical staining of MCM family members using data from the Human Protein Atlas database to evaluate the protein levels of MCMs in LUSC. Figure 2 shows that LUSC tissues expressed higher protein levels of MCM3/4/5, compared to the normal lung tissues (not detected versus medium, respectively) (Figure 2C–2E). Similar findings were observed for the protein expression levels of MCM2/6/7/10 (not detected versus high, medium versus high, not detected versus high, and low versus medium, respectively) (Figure 2F, 2I). These findings are consistent with our previous findings regarding the mRNA expression of the MCM family genes. Moreover, MCM1 was not expressed in normal lung tissues and showed medium expression in LUSC tissues (Figure 2A), whereas MCM9 showed low expression in normal tissues and medium expression in LUSC tissues (Figure 2H).

Figure 2. Typical immunohistochemistry images of minichromosome maintenance (MCM) family members. (A–I) Comparison of the expression of MCM1–10 in lung squamous cell carcinoma (LUSC) tissues with those in non-cancerous tissue (100×) using the data from the Human Protein Atlas database.

Correlation between MCM mRNA expression and clinicopathological parameters in patients with LUSC

The association between the mRNA expression of the ten MCM genes and lymph node metastases in LUSC was examined using the UALCAN database. At all stages of lymph node metastasis, the MCM2/4/6/7/10 mRNA expression in LUSC tissues were higher than those in normal tissues. Additionally, the mRNA expression of MCM3/5/8/9 tended to be higher in tumors with N0–N2 stage lymph node metastasis. At the N0–N3 stage of lymph node metastasis, MCM4/6/7/10 showed the lowest expression in N3 tumors, which significantly correlated with patient prognosis. In contrast, the expression of MCM2 was highest at N3 stage (Figure 3A). Therefore, the mRNA expression of MCM family members were significantly correlated with individual cancer stages of LUSC. The expression of eight MCM genes (excluding MCM1 and MCM9) was significantly higher in the tumor stage 1–4 subgroups than that in normal lung tissues. However, the expression of MCM9 did not differ between stage 4 LUSC tissue and normal lung tissue. In addition, MCM3/6/8/10 mRNA expression was lower in tumor stage 4 than in tumor stages 1 and 3. These findings suggest that MCMs contribute to the development of LUSC (Figure 3B).

Figure 3. Correlation between clinical pathology and minichromosome maintenance (MCM) mRNA expression determined using UALCAN. (A) Association of lymph node metastases in patients with lung squamous cell carcinoma (LUSC) with MCM family mRNA expression. (B) Association of pathological stage of patients with LUSC with MCM family mRNA expression. *p < 0.05.

Prognostic value of MCM gene family in patients with LUSC

Based on mRNA expression in patients with LUSC, the prognostic value of the MCM family was assessed using the Kaplan–Meier plotter database. The clinical prognosis was determined through survival analysis considering post-progression survival (PPS), recurrence-free survival (RFS), and overall survival (OS). We found that higher mRNA expression of MCM3 [OS: hazard ratio (HR) = 0.71 (0.53–0.96), P = 0.024], MCM4 [OS: HR = 0.68 (0.5–0.93), P = 0.016], MCM5 [OS: HR = 0.72 (0.54–0.97), P = 0.029], MCM6 [OS: HR = 0.68 (0.51–0.91), P = 0.0095], MCM7 [OS: HR = 0.62 (0.46–0.82), P < 0.001], and MCM8 [OS: HR = 0.74 (0.56–0.97), P = 0.031] was associated with favorable OS in patients with LUSC and may be suitable targets for improving patient survival and prognosis (Figure 4A). Moreover, increased expression of MCM3 [RFS: HR = 1.93 (1.04–3.75), P= 0.033], MCM7 [RFS: HR = 1.82 (1.09–3.04), P = 0.021], MCM8 [RFS: HR = 2.16 (1.3–3.57), P = 0.002], and MCM10 [RFS: HR = 1.95 (1.09–3.5), P = 0.022] was significantly associated with poorer RFS in patients with LUSC (Figure 4B). Upregulation of MCM1 [PPS: HR = 1.29 (1–1.66), P = 0.049], MCM2 [PPS: HR = 1.51 (1.17–1.94), P = 0.001], MCM3 [PPS: HR = 1.36 (1.05–1.77), P = 0.020], MCM4 [PPS: HR = 1.62 (1.25–2.09), P < 0.001], MCM5 [PPS: HR = 1.45 (1.11–1.9), P = 0.007], MCM7 [PPS: HR = 1.54 (1.2–1.98), P < 0.001], and MCM9 [PPS: HR = 1.59 (1.22–2.08), P < 0.001] was significantly associated with longer PPS (Figure 4C). The mRNA expression of other MCMs had no significant effect on OS, RFS, or PPS in patients with LUSC (Figure 4). Members of the MCMs family had inconsistent results for OS, RFS, and PPS in patients with LUSC. This situation may be due to the small number of patient samples, and further studies are needed. Among the MCMs, MCM3 and MCM7 showed the most significant correlation with the clinical prognosis of LUSC, indicating the potential of MCM3 and MCM7 as prognostic markers for LUSC.

Figure 4. Prognostic value of MCM family members in lung squamous cell carcinoma (LUSC). (A–C) Association of mRNA expression of MCM family members with overall survival (OS), progression-free survival (RFS), and post-progression survival (PPS) in LUSC using Kaplan–Meier plotter database.

We further analyzed the prognostic value of the MCM family members at different clinical stages of LUSC using Kaplan–Meier plotter database (Table 1). Overexpression of MCM7 [HR = 2.44 (1.05–5.68), P = 0.033] and MCM10 [HR = 2.94 (1.17–7.37), P = 0.016] was significantly associated with a poorer RFS in patients with stage I LUSC. The mRNA expression of MCM2 [HR = 7.58 (1.61–35.66), P = 0.0028], MCM3 [P = 0.038], MCM4 [HR = 4.12 (1.2–14.1), P = 0.015], MCM5 [HR = 4.81 (1.04–22.32), P = 0.027), MCM8 (HR = 3.63 (1.09–12.09), P = 0.024], and MCM9 [HR = 6.2 (0.8–48.12), P = 0.046] was significantly associated with poorer RFS in patients with stage III LUSC. These results indicate that some MCM family members can be used as prognostic factors in LUSC, particularly for RFS prediction in patients diagnosed at an advanced stage of LUSC.

Genetic alterations and functional enrichment analysis of MCM family members

Genetic alterations are well-known to be major factors influencing cancer development. We determined the methylation levels of MCM genes in patients with LUSC using the UALCAN database. The DNA methylation levels of MCM3/5/6/8 were significantly lower in LUSC samples than in normal tissues, whereas those of MCM1/2/4/7/10 were significantly higher. Other MCM family members showed no significant differences between normal and cancer tissues (Figure 5). Variations in the methylation levels of MCM genes may have been caused by differences in their mRNA expression. Thus, DNA methylation-targeting drugs may be useful for treating cancer.

Figure 5. DNA methylation levels of minichromosome maintenance (MCM) family genes in lung squamous cell carcinoma (LUSC). (A–J) DNA methylation change of MCM1–10 in LUSC investigated using the UALCAN database. *p < 0.05 compared with control.

In addition, we used the cBioPortal database to conduct a series of studies to verify the status of genetic alterations in the MCM family. We found that 80 of 178 samples (45%) contained alterations in MCM genes, with MCM2 showing the largest proportion of alterations (15%). The two most prevalent genetic changes in the MCM family were mRNA upregulation and amplification (Figure 6A). Next, we identified co-expressed genes with threshold values of |log2 fold-change| ≥ 0.45 and P < 0.05 using the cBioPortal database (Supplementary Table 1) and used Cytoscape v.3.9.0 to generate a map of the co-expression networks of key genes related to the MCM family (Figure 6B).

Figure 6. Genetic alterations and pathway enrichment analysis of minichromosome maintenance (MCM) family in LUSC. (A) Summary of mutation rates in each MCM member in LUSC. (B) Protein–protein interaction (PPI) network of the interaction partners of MCM family members built using cBioPortal and Cytoscape. (C–F) KEGG enrichment pathway analysis of molecular functions, biological processes, and cellular components of co-expressed genes.

The Metascape database was utilized for Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis of co-expressed genes. Hematopoietic cell lineage, cell adhesion molecules, neuroactive ligand-receptor interaction, and cytokine–cytokine receptor interaction were explored for co-expressed genes using KEGG pathway analysis (Figure 6C). Molecular function analysis indicated that the genes were primarily involved in lipid antigen binding, ligand-gated ion channel activity, and immune receptor activity (Figure 6D). Biological process analysis indicated that these genes were mainly involved in antigen processing and presentation and behavioral fear responses (Figure 6E). Cellular component analysis revealed that these genes were frequently associated with clathrin-coated vesicle membranes and neuronal cell bodies (Figure 6F). These results indicate that the MCM family is involved in antigen processing and presentation, immune receptor activity, and cytokine-cytokine receptor interaction in LUSC, which may impact immune infiltration of the tumor microenvironment (TME) in LUSC tissues.

Association of expression of MCM family members with immune infiltration in LUSC

The correlation between the MCM gene family and immune cell infiltration was investigated using the TIMER database. The results showed that expression of MCM1 mRNA was significantly associated with infiltration of CD8+ T cells [correlation coefficient (cor) = −0.205, P < 0.05] and CD4+ T cells (cor = 0.260, P < 0.05). Expression of MCM2 was significantly associated with B-cell (cor = 0.112, P < 0.05) and macrophage (cor = − 0.978, P < 0.05) infiltration, while MCM3 expression was significantly associated with infiltration of CD4+ T cells (cor = 0.178, P < 0.05) and macrophages (cor = −0.176, P < 0.05). Additionally, MCM4 expression was significantly associated with infiltration of CD4+ T cells (cor = 0.143, P < 0.05) and macrophages (cor = −0.093, P < 0.05), while expression of MCM5 was significantly associated with infiltration of CD8+ T cells (cor = −0.143, P < 0.05), CD4+ T cells (cor = 0.152, P < 0.05), and macrophages (cor = −0.110, P < 0.05). MCM7 expression was significantly associated with macrophage infiltration (cor = −0.179, P < 0.05), and expression of MCM8 was significantly associated with infiltration of CD4+ T cells (cor = 0.139, P < 0.05), while MCM9 expression was significantly associated with infiltration of CD4+ T cells (cor = 0.203, P < 0.05). Finally, MCM10 expression was significantly associated with macrophage infiltration (cor = −0.185, P < 0.05) (Figure 7). These results suggest that MCM family members affect the immune response in the TME of LUSC.

Figure 7. Relationship between mRNA expression of minichromosome maintenance (MCM) family and immune cell infiltration. (A–C) Analysis of association of MCM1–10 mRNA expression with the level of immune cell infiltration in lung squamous cell carcinoma (LUSC) using TIMER database.
(D–J) Analysis of association of MCM1–10 mRNA expression with the level of immune cell infiltration in lung squamous cell carcinoma (LUSC) using TIMER database.

Analysis using the TIMER database showed that expression of the MCMs was highly correlated with the signature marker genes of various immune cells in LUSC (Table 2). Notably, MCM1 levels were significantly associated with CD8+ T cells, T cells, M1 and M2 macrophages, T follicular helper (Tfh) cells, Th17 cells, and neutrophils. Expression of MCM2 was significantly correlated with tumor-associated macrophages (TAMs), Th17 cells, and regulatory T cells (Tregs). MCM3 expression was significantly associated with TAMs, M1 macrophages, Tfh, and Th17 cells in LUSC, while MCM4 expression exhibited a strong correlation with B cells, T cells, TAMs, M1 macrophages, dendritic cells, Tfh cells, and Th17 cells. MCM5 expression was significantly associated with most signature marker genes of T cells, TAMs, M1 macrophages, Tfh cells, and Th17 cells. Furthermore, MCM6 expression was associated with B cells, M1 macrophages, Tfh cells, and Th17 cells. MCM7 expression showed high correlations with B cells, T cells, TAMs, M1 macrophages, Tfh cells, and Th17 cells. The mRNA expression of MCM8 was strongly associated with CD8+ T cells, B cells, T cells, TAMs, Th2 cells, Tfh cells, and Th17 cells. The mRNA expression of MCM9 was correlated with M1 macrophages, Tfh cells, Th17 cells, and Tregs. Moreover, B cells, T cells, TAMs, M1 and M2 macrophages, neutrophils, dendritic cells, Th2 cells, and Tfh cells showed favorable correlations with MCM10 expression. Taken together, these findings reveal important relationships between MCM family members and immune-infiltrating cells in LUSC and suggest that these proteins play important roles in the TME of LUSC.

TIMER

Based on The Cancer Genome Atlas database, the TIMER tool is a comprehensive resource for evaluating immune cell infiltration and the clinical outcomes of 10,897 tumors from 32 different cancer types. This database enables analyses of the correlations between genes and immune-infiltrating cells, comparisons of gene expression in tumors and normal tissues in various malignancies, and survival analysis, among other functions [36]. We investigated the mRNA expression of MCM family members in various malignancies or specific cancer subtypes from TIMER, which contains 501 LUSC samples and 51 normal lung samples; the log2 (transcripts per million) were applied to convert the data to a log-scale. We also examined the relationship between the MCM family and immune cell infiltration using the TIMER database.

GEPIA2

The web-based tool GEPIA2 provides vital interactive and customizable functions, including patient survival analysis, correlation analysis, and differential expression analysis [37]. 338 normal lung samples and 486 primary LUSC samples were included in the GEPIA2 dataset. The GEPIA2 datasets were utilized to compare the expression of different MCMs in LUSC and normal tissues, and p < 0.05 was considered to indicate statistically significant results.

Human protein atlas

Human Protein Atlas is an online database containing information on protein expression in various cancer types based on immunohistochemistry [38]. We used immunohistochemical images to evaluate the protein levels of several MCM members between LUSC tumors and normal lung tissues.

UALCAN

UALCAN is an interactive network resource based on The Cancer Genome Atlas datasets that allows users to compare information on DNA methylation and mRNA expression in human malignancies [39]. To examine the relationships between the mRNA expression of various MCMs and clinicopathological features in 856 LUSC samples and 52 normal lung samples, we used the “individual cancer stages” model and “nodal metastatic status” model. Using the UALCAN database, we further predicted DNA methylation alterations in MCM family members from 370 LUSC cases and 42 normal lung tissues. Student’s t-test was applied, and p < 0.05 was considered to indicate statistically significant results.

Kaplan–Meier plotter database

The Kaplan–Meier plotter database (https://kmplot.com) was used to analyze the prognostic value of the MCM family in LUSC. The mRNA expression value of the auto-selected best cutoff was used to split the LUSC samples into high and low expression groups to investigate the relationship between MCM expression and OS, RFS, and PPS for 501 patients with LUSC. Using the Kaplan–Meier plotter database, we further examined the prognostic value of MCM family members in patients with various clinical stages of LUSC. A p-value less than 0.05 was considered to indicate statistically significant results.

cBioPortal

cBioPortal (http://cbioportal.org), an open comprehensive platform, can be used to analyze multidimensional cancer genomics and clinical data. We examined the genomic map of the MCM family members, which contains information on mutations and mRNA expression, in 178 LUSC samples. The threshold of the |log2 fold-change| was 0.45, and the p-value cutoff was set at 0.05.

STRING

The STRING database (https://cn.string-db.org/) was used to assess correlations involving MCM genes.

Cytoscape

We integrated functionally 186 co-expressed MCM family members selected using the cBioPortal (molecule names are provided in Supplementary Table 1). The node size was determined as the degree values between interacting proteins.

Metascape

The web-based analysis toolkit Metascape is comprehensive, efficient, customizable, and interactive [40]. Using this dataset, we conducted GO and KEGG pathway enrichment analyses for genes co-expressing with MCM members.

Data availability statement

The original contributions presented in the study are included in the Supplementary Materials, further inquiries can be directed to the corresponding authors.