MCM10: An effective treatment target and a prognostic biomarker in patients with uterine corpus endometrial carcinoma

Abstract Molecular profiling has been applied for uterine corpus endometrial carcinoma (UCEC) management for many years. The aim of this study was to explore the role of MCM10 in UCEC and construct its overall survival (OS) prediction models. Data from TCGA, GEO, cbioPotal and COSMIC databases and the methods, such as GO, KEGG, GSEA, ssGSEA and PPI, were employed to bioinformatically detect the effects of MCM10 on UCEC. RT‐PCR, Western blot and immunohistochemistry were used to validate the effects of MCM10 on UCEC. Based on Cox regression analysis using the data from TCGA and our clinical data, two OS prediction models for UCEC were established. Finally, the effects of MCM10 on UCEC were detected in vitro. Our study revealed that MCM10 was variated and overexpressed in UCEC tissue and involved in DNA replication, cell cycle, DNA repair and immune microenvironment in UCEC. Moreover, silencing MCM10 significantly inhibited the proliferation of UCEC cells in vitro. Importantly, based on MCM10 expression and clinical features, the OS prediction models were constructed with good accuracy. MCM10 could be an effective treatment target and a prognostic biomarker for UCEC patients. The OS prediction models might help establish the strategies of follow‐up and treatment for UCEC patients.

countries because of poor medical care. 2 This phenomenon poses a serious threat to women's health globally. Although most of the patients with early phase UCEC could be cured by hysterectomy and adjuvant radiotherapy, women with advanced UCEC still have a poor prognosis, with five-year survival rates of approximating 48% in stage III and 15% in stage IV. 3 The combination of carboplatin and paclitaxel, the first line of advanced UCEC, only results in 13 months of progress-free survival. 4 Unfortunately, within the last three decades, there were only a few drugs approved for use in advanced UCEC by the US Food Drug Administration (FDA). Chemotherapy remains a standard management for patients with advanced or recurrent UCEC. 4,5 Molecular profiling has been used to manage these advanced and recurrent UCEC patients for many years. For example, progesterone receptor and oestrogen receptor status have been used to evaluate whether patients can be treated with hormone therapy.
However, even in some patients with positive progesterone receptor and oestrogen receptor, single-agent aromatase inhibitors only achieved 10% response rate. 2 The use of mTOR inhibitor yielded a response rate of merely 10%. 6 Hence, it is still urgent to find new targets and further understand its molecular biology in UCEC, and these efforts will contribute to effective targeted chemotherapeutic strategies in the future.
Minichromosome maintenance 10 replication initiation factor (MCM10) gene is a protein-coding gene located on chromosome region 10p13. MCM10 was first found to involve in the initiation process of DNA replication in yeast. 7 It has been revealed that MCM10 is highly expressed in bone marrow, lymph node and appendix, but relatively low in endometrium tissue. 8 In recent years, MCM10 has been proved to promote the DNA elongation process through inhibiting the activity of Cdc45/MCM2-7/GINS complex and ensuring the stability of replication fork. 9,10 These properties allow MCM10 to play an important role in cell proliferation, and even in cellular immortalization. Overexpression of MCM10 has been found in various cancers, including ovarian cancer, 11 breast cancer 12 and prostate cancer 13 and is associated with poor prognosis. A pan-cancer analysis based on GEPIA database showed that MCM10 mRNA expression was overexpressed in UCEC as compared with normal endometrium. 11 Considering the close relationship between MCM10 and various malignant tumours, we believe that MCM10 may also be associated with the development of UCEC and can be an effective prognosis biomarker.
In this study, data from breast cancer, cervical cancer, uterine sarcoma and UCEC in The Cancer Genome Atlas (TCGA) database were used to screen novel biomarkers, and MCM10 was identified to be an effective indicator to predict UCEC prognosis. Subsequently, based on the bioinformatic analysis, we explored the potential func-

| The screening of biomarkers predicting the prognosis of UCEC
The data sets of breast cancer (BRCA), cervical squamous cell carcinoma and endocervical adenocarcinoma (CESC), uterine carcinosarcoma (UCS) and UCEC from TCGA database were employed to screen the biomarkers predicting the prognosis of UCEC. The R package 'DESeq2' was used to obtain four differentially expressed genes (DEGs) between tumour and normal tissue (tumour vs. normal) based on the data sets of BRCA, CESC, UCS and UCEC, respectively.
The Log 2 Fold Change (LogFC) > 3 and adjusted p-value <0.05 were considered as the cut-off of DEGs. After taking the intersection of the four DEGs, the top 20% overexpression genes in UCEC were selected to be further screened by Kaplan-Meier analysis in UCEC.
Finally, the gene, MCM10, has not been reported to be associated with UCEC were selected in this study.
The differential expression data of MCM10 in unpaired and paired samples are RNA-seq data from TCGA UCEC project and GTEx database. The data were uniformed by the Toil process. 14 All final analyses based on TCGA were performed using data in TPM format. The differential analysis data for MCM10 in UCEC based on the data set GSE17025 was employed to further validate that MCM10 mRNA was overexpressed in UCEC tissues.

| Single-gene differential analysis and correlation analysis of MCM10
Single-gene differential analysis and single-gene correlation analy- were considered as the cut-off of the DEGs. STRING database was used to visualize the analysis of DEGs, 16 and protein-protein interaction (PPI) of DEGs was performed using Cytoscape software network analysis, and then the HUB genes were identified using the MCODE plugin. Finally, the genes were sorted by 'Pearson value' in descending order based on single-gene correlation analysis, and the genes whose correlations were in the top 50 were extracted.

| Functional enrichment analysis
Gene set enrichment analysis (GSEA) was used to explore potential signalling pathways based on the differential expression

| Analysis of the correlation between MCM10 expression and prognosis of patients with UCEC
The survival data of UCEC patients from TCGA database were used to analyse the correlation between the mRNA expression of various genes (including HOXB13, PITX1, MYBL2, IGF2BP1, CDC20, NMU,

| Specimens
The present study was approved by the Ethics Committee of the  Table 1. The median follow-up for patients with UCEC was 43 months, ranging between 7 and 107 months.

| Cell culture and stably transfected cell line development
The human UCEC cell lines Ishikawa and HEC-1-A (iCell Bioscience Inc.) were cultured at 37 C in a humidified atmosphere with 5% into six-well plates (300,000 cells/well). 1 mL medium containing the above lentiviral was added into each well. The cells were selected in medium containing puromycin (2 μg/mL) and maintained in medium containing puromycin (1 μg/mL). Puromycin selection was continued for 1 week prior to harvesting cells for downstream analysis.

| Colony formation assay
The indicated cells were seeded onto 6-well plates at a density of 100 cells/well. 10 days later, the formation of typical clone of the cells was observed. The cells were fixed with methanol and stained with 10% Giemsa (Biotopped). The number of visible colonies was counted to evaluate the colony formation ability of cells. 21

| Immunofluorescence
The indicated cells were seeded on six-well plates at 2 × 10 6 cells/ well. Standard immunofluorescence procedures were carried out as a previous study. 22 Rabbit anti-MKI67 (Servicebio) antibody was used as the primary antibody. Goat anti-Rabbit IgG (H + L) conjugated with Cy5 (Servicebio) was used as the secondary antibody.
The cover slips were observed using a fluorescence microscope (Olympus). Blue represented the nucleus stained with DAPI, and red represented MKI67 protein.

| Statistical analysis
ssGSEA was used for the algorithm of immune infiltration. Statistical analyses were performed with SPSS 23.0 or R version 3.6.3, the differences among the groups were compared using one-factor analysis of variance (anova) followed by Dunnett's post hoc test, Kruskal-Wallis test or student's t test. Crosstab data were compared by Chi-squared or Fisher's test. Differences in p < 0.05 were considered statistically significant. All results were repeated three times independently.
These genes were then selected for survival analysis in UCEC.
Finally, MYBL2, IGF2BP1, NMU, UBE2C and MCM10 were found to be associated with the OS of UCEC patients (p < 0.05) ( Figure 1A and Figure S1). To the best of our knowledge, only MCM10 has not been reported to be associated with the development and prognosis of UCEC. Thus, we selected MCM10 as a new target to explore its relationship with UCEC. Based on TCGA database, we found that the expression of MCM10 mRNA was upregulated in UCEC tissues, both compared with normal tissues ( Figure 1B) and adjacent tissues ( Figure 1C). Moreover, the result from GSE17025 database also testified that MCM10 mRNA was overexpressed in UCEC tissues as compared to that in normal control tissues ( Figure 1D).

| Variation of MCM10 in UCEC
The results from the cBioPortal database showed that 26.72% (136/509) of UCEC patients had variation of MCMs ( Figure 1E), of which MCM10 accounted for 5.7% (29/509), and mutation was the most frequent variation ( Figure 1F). Therefore, COSMIC database was used to further explore the types of mutation. The results showed that missense substitution occurred in 53.12% of the UCEC samples and synonymous substitution occurred in 37.50% of the UCEC samples ( Figure 1G). Furthermore, the base substitutions were mainly G > A (42.31%), C > T (34.62%) and A > G (19.23%) ( Figure 1H).

| Enrichment analysis of the MCM10 expression phenotype
Based on TCGA database, a total of 3966 DEGs (|LogFC| > 1, adjusted p-value <0.05) were identified between UCEC samples with high expression and low expression of MCM10 (Table S7). In this regard, the samples with low MCM10 expression were considered as controls. As shown in the volcano plot (Figure 2A Figure 2D). More GSEA results were shown in Table S9.

| Analysis of genes associated with MCM10 in UCEC
As shown in Figure 2E, 63 HUB genes related to MCM10 in UCEC were identified in the PPI network using MCODE plugin. The correlation between MCM10 mRNA expression and the top 50 of these HUB genes was presented in Figure 2F. Moreover, the correlation between MCM10 mRNA expression and MCM family genes, including MCM2, MCM3, MCM4, MCM5, MCM6, MCM7, MCM8 and MCM9, was shown in Figure 2G. Finally, we found that most of MCMs mRNA expression was positively correlated with MKI67 mRNA expression (r > 0.7, p < 0.01) ( Figure 2H).

| The effects of MCM10 changes on immune cell infiltration in UCEC
To detect the effects of MCM10 changes on the tumour microenvironment, immune infiltration analysis was performed using the ssGSEA method. As shown in Figure S2A and Table S10, the pro-

portion of T help2 (Th2) cells was significantly higher in MCM10
high-expression group than that in the low-expression group. The

| Analysis of the association between MCM10 expression and the clinical features of patients with UCEC
To further demonstrate the effect of MCM10 expression on UCEC development in patients, the association between MCM10 expression and clinical features of patients with UCEC was evaluated using TCGA database and the clinical data from the Second Hospital of Jilin University. The clinical baseline information tables of TCGA database (Table S11) and our clinical data (Table 1) Figure 3G) or menopause status of the patients ( Figure 3H). More specifically, the mRNA expressions of MCM10 were higher in the groups of age > 60, tumour invasion > = 50%, serous type, grade 2 (G2), grade 3 (G3), stage III and IV and part response (PR) compared with that in the control groups (p < 0.05). Furthermore, the logistics regression analysis showed that age > 60, serous type, G3 and stage III and IV were risk factors of MCM10 mRNA high expression in UCEC ( Figure 3I).

The analysis based on our clinical data showed that MCM10
protein expression significantly (p < 0.05) increased in UCEC tissues as compared with that in normal or adjacent tissues Furthermore, the logistics regression analysis showed that tumour F I G U R E 2 Gene function analysis of MCM10 based on TCGA database. (A) Volcano plot of the DEGs analysis, and the top ten genes with the greatest differences were marked; (B) KEGG and GO enrichment analysis of DEGs (|LogFC| > 1, adjusted p-value <0.05); (C, D) GSEA enrichment analysis of DEGs; (E) PPI analysis of DEGs; (F) The top 50 genes correlated with MCM10 was showed in the heat map; (G) the correlation between MCMs and MCM10 was showed in the heat map; (H) the correlation between MCMs and MKI67 was shown in the heat map. **p < 0.01, ***p < 0.001. invasion > = 50% and stage III and IV were risk factors of MCM10 protein high expression in UCEC ( Figure 4S).

| Construction, evaluation and visualization of OS prediction models based on the results of multivariate Cox regression in UCEC patients
As shown in Figure S3, the univariate Cox regression analysis identified that age > 60, serous type, mixed histological type, G3, stage III, stage IV, R2, PR and MCM10 mRNA expression levels were significant risk factors for the OS of patients with UCEC in TCGA database (p < 0.05).
Radiation therapy was a prevention factor for the OS of UCEC patients (p < 0.05). Based on the results of univariate Cox regression analysis, we found that age > 60, mixed histological type, stage III, stage IV, PR and MCM10 mRNA expression levels were independent risk factors for the OS of patients with UCEC in TCGA database (p < 0.05) ( Figure 5A). As shown in Figure 5B, the OS prediction model included MCM10 mRNA expression level, histological type, clinical stage and primary therapy outcome, which was visualized by the nomogram. The calibration plot of the model showed satisfactory agreement between the prognostic prediction and actual observation ( Figure 5C). Figure S4, the univariate Cox regression analysis identified that age > =55, serous type, mixed histological type, G2, G3, stage III and stage IV, lymphatic metastasis, PR and MCM10 protein expression level were significant risk factors for the OS of patients with UCEC in our clinical data (p < 0.05). Based on the results of multivariate Cox regression analysis, we found that mixed histological type, stage III and stage IV, PR and MCM10 protein expression levels were independent risk factors for the OS of patients with UCEC (p < 0.05) ( Figure 6A). As shown in Figure 6B, the OS prediction model included MCM10 protein expression level, histological type, clinical stage and primary therapy outcome, which was visualized by the nomogram. The calibration plot of the model showed satisfactory agreement between the prognostic prediction and actual observation ( Figure 6C).

| Proliferation ability of UCEC cells is inhibited by downregulating MCM10 in vitro
The results of RT-PCR ( Figure 7A) and Western blot showed that the expression of MCM10 in Ishikawa ( Figure 7B) and HEC-1-A ( Figure 7C) cells was significantly downregulated after being transfected with shMCM10 (p < 0.05). As shown in Figure 7D, from day 2, compared with the Mock and shScramble groups, the cell proliferation of Ishikawa and HEC-1-A cells in the shMCM10 groups was remarkably lower (p < 0.05). Moreover, the colony ability of Ishikawa and HEC-1-A cells in shMCM10 groups was significantly (p < 0.05) attenuated compared to the Mock and shScramble groups ( Figure 7E).
Importantly, the expression of MKI67 significantly decreased in shMCM10 groups as compared with the MOCK and shScramble groups ( Figure 7F).

| DISCUSS ION
MCMs have been identified to contribute to tumorigenesis and tumour cell proliferation in various cancers. 23,24 As a member of MCMs, research on MCM10 has been increasing in recent years.
Just like other MCMs, the variation of MCM10 gene has been found in many cancers, such as breast cancer, ovarian cancer and gastric cancer. 10,11,25 Unlike in ovarian cancer where MCM10 gene variation is mainly concentrated on amplification, 11 the variation of MCM10 is mainly concentrated on mutations in UCEC, which might be a potential mechanism of our finding that both of MCM10 mRNA and protein were overexpressed in UCEC tissue. However, to the best of our knowledge, no studies have explored the role of MCM10 in the development of UCEC. In our study, the expression level of MCM10 was found to be related to multiple clinical features that were associated with the development of UCEC, including age, histological type, histologic grade, tumour invasion, clinical stage, primary therapy outcome and lymphatic metastasis. Taken together, these data above clearly demonstrated that MCM10 changed in the genome, mRNA and protein levels of UCEC, but the effects of these changes on UCEC development still remains unclear.
In order to further explore the effects of MCM10 changes on UCEC, DEGs analysis based on TCGA database was used to enrich the underlying pathways. Subsequently, the results of gene function enrichment were accordant with the previous studies, which indicated that MCM10 positively regulated replication initiation, 26 cell cycle 27 and DNA repair. 28 It is well known that the unlimited DNA replication is the basic characteristic of tumour cell proliferation, which closely depends on the cell cycle process. 29 Therefore, high expression of MCM10 might contribute to the cell proliferation in UCEC, and this hypothesis was further verified by our results that the mRNA expression levels of most of MCMs, including MCM10, F I G U R E 4 Correlations between MCM10 protein expression and clinical features were validated by clinical samples. The significant changes of MCM10 protein expression in (A, G, H, I) Normal tissues, adjacent tissues of UCEC, G1 UCEC tissues and G2 and G3 UCEC tissues, (B, J) UCEC tissues of patients with FIGO I, FIGO II, FIGO III and FIGO IV, (C, K) UCEC tissues of patients with invasion <50% and invasion > = 50%, (D, L) UCEC tissue of patients with complete response (CR) and partial response (PR), and (E, M) UCEC tissues of patients with lymphatic metastasis and no lymphatic metastasis; (F, N) MCM10 expression was positively correlated with MKI67; the expression of MCM10 protein in (P) UCEC tissues of Endometrioid, Serous, and Mixed histological types, (Q) UCEC tissues of patients with diabetes and no diabetes and (R) UCEC tissues of patients with menopause status and pre-menopausal status; (S) the risk factors for MCM10 protein high expression were shown in the forest plot based on logistics regression analysis; *p < 0.05, **p < 0.01, and ***p < 0.001.
were positively associated with the expression levels of MKI67 mRNA in UCEC. MKI67 is a nuclear protein expressed in proliferating mammalian cell, which is one of the most widely used markers of proliferation in oncology and widely used as a proliferation indicator in the clinic. [30][31][32][33] In UCEC, MKI67 has been found to be related to the proliferation and invasion of UCEC cells. Thus, the regulation effect of MCM10 on UCEC could be revealed by its correlation with MKI67. Furthermore, our in vitro study indicated that blocking MCM10 was an effective strategy to inhibit UCEC cell proliferation and MKI67 expression, which implied that MCM10 targeting treatment might be beneficial for UCEC patients. However, it still needs more experiments to testify whether these effects were mediated In this study, we found that the results obtained by using TCGA database to analyse the correlation between MCM10 mRNA and UCEC clinical characteristics were inconsistent with those obtained by using immunohistochemistry to analyse the correlation between In recent years, based on tumour microenvironment-and immune-related genes, several models predicting the prognosis of endometrial cancer have been constructed. 41,42 Considering that MCM10 was identified as a tumour immune microenvironmentrelated gene and an independent risk factor for the OS of UCEC, TCGA data-based prognosis prediction model for UCEC, incorporating MCM10 expression and some other independent risk factors, was firstly established in this study. Importantly, this model was then validated in our clinical data-based prognosis prediction model.
According to the expression of MCM10 and the independent risk factors from conventional clinical data, both models could be used to predict the individual 4-, 6-and 8-year OS probability. This novel strategy could be valuable in designing individualized long-term follow-up protocols and therapeutic regimes.

| Conclusions
In conclusion, based on public databases and our clinical data, we

CO N FLI C T O F I NTE R E S T S TATE M E NT
The authors declare that they have no competing interests.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data sets generated during and/or analysis during the current study are available from the corresponding author on reasonable request.

PATI E NT CO N S E NT S TATE M E NT
The patients were informed and agreed to participate in this study.