Overexpression of ASPM, CDC20, and TTK Confer a Poorer Prognosis in Breast Cancer Identified by Gene Co-expression Network Analysis

Abstract

Breast cancer is one of the most common malignancies among females, and its prognosis is affected by a complex network of gene interactions. In this study, we constructed free-scale gene co-expression networks using weighted gene co-expression network analysis (WGCNA). The gene expression profiles of GSE25055 were downloaded from the Gene Expression Omnibus (GEO) database to identify potential biomarkers associated with breast cancer progression. GSE42568 was downloaded for validation. A total of 9 modules were established via the average linkage hierarchical clustering. We identified 3 hub genes (ASPM, CDC20, and TTK) in the significant module (R ² = 0.52), which were significantly correlated with poor prognosis both in test and validation datasets. In the datasets GSE25055 and GSE42568, higher expression levels of ASPM, CDC20, and TTK correlated with advanced tumor grades. Immunohistochemistry data from the Human Protein Atlas also demonstrated that their protein levels were higher in tumor samples. According to gene set enrichment analysis, 4 commonly enriched pathways were identified: cell cycle pathway, DNA replication pathway, homologous recombination pathway, and P53 signaling pathway. In addition, strong correlations were found among their expression levels. In conclusion, our WGCNA analysis identified candidate prognostic biomarkers for further basic and clinical researches.

Keywords: ASPM; CDC20; TTK; WGCNA; breast cancer; prognosis.

Figure 1

Clustering dendrogram and determination of soft-thresholding power in the WGCNA. (A) Clustering dendrogram of 302 samples. (B) Analysis of the scale-free fit index for various soft-thresholding powers (β). (C) Analysis of the mean connectivity for various soft-thresholding powers (β). We choose the lowest β that results in approximate scale free topology. (D) Checking the scale free topology when β = 8. The x-axis shows the logarithm of whole network connectivity, y-axis shows the logarithm of the corresponding frequency distribution. On this plot the distribution approximately follows a straight line, which is referred to as approximately scale-free topology.

Figure 2

Identification of modules associated with the clinical traits of breast cancer. (A) Dendrogram of all differentially expressed genes clustered based on a dissimilarity measure (1-TOM). The color band provides a simple visual comparison of module assignments. The color band shows the results from the automatic single block analysis. (B) Heatmap of the correlation between module eigengenes and clinical traits of breast cancer. (C) Distribution of average gene significance and errors in the modules associated with tumor grades of breast cancer.

Figure 3

Protein-protein network and gene enrichment analysis of brown module genes. (A) Protein-protein network, the red nodes represent hub genes in the module. (B) Gene enrichment analysis.

Figure 4

Relapse free survival (RFS) and overall survival (OS) of the 3 hub genes in breast cancer in dataset GSE25055 and GSE42568. The patients were stratified into high-level group and low-level group according to median expression. (A) RFS of ASPM in GSE25055. (B) RFS of CDC20 GSE25055. (C) RFS of TTK GSE25055. (D) RFS of ASPM in GSE42568. (E) RFS of CDC20 in GSE42568. (F) RFS of TTK in GSE42568. (G) OS of ASPM in GSE42568. (H) OS of CDC20 in GSE42568. (I) OS of TTK in GSE42568.

Figure 5

Overall survival (OS) and relapse free survival (RFS) of the 3 hub genes in breast cancer based on Kaplan Meier-plotter. The patients were stratified into high-level and low-level groups according to median expression. (A) RFS of ASPM. (B) OS of ASPM. (C) RFS of CDC20. (D) OS of CDC20. (E) RFS of TTK. (F) OS of TTK.

Figure 6

Validation of ASPM, CDC20 and TTK. (A–C) Expression of hub genes in different tumor grades based on GSE25055. (E–F) Expression of hub genes in different tumor grades based on GSE42568. ^***P < 0.001; ^****P < 0.0001. Student's t-tests were used to evaluate the statistical significance of differences.

Figure 7

Expression levels of ASPM, CDC20 and TTK. (A) ASPM expression and breast cancer subtypes. (B) CDC20 expression and breast cancer subtypes. (C) TTK expression and breast cancer subtypes. (D) ASPM expression and tumor stages. (E) CDC20 expression and tumor stages. (F) TTK expression and tumor stages. (G) ASPM expression and tumor sizes. (H) CDC20 expression and tumor sizes. (I) TTK expression and tumor sizes. (J) ASPM expression and lymph node status. (K) CDC20 expression and lymph node status. (L) TTK expression and lymph node status. ^*P < 0.05; ^**P < 0.01; ^***P < 0.001; ^****P < 0.0001. One-way analysis of variance (ANOVA) and two-tailed Student's t-tests were used to evaluate the statistical significance of differences.

Figure 8

Immunohistochemistry of the six hub genes based on the Human Protein Atlas. (A) Protein levels of ASPM in normal tissues (https://www.proteinatlas.org/ENSG00000066279-ASPM/tissue/breast#img). (B) Protein levels of ASPM in tumor tissues (https://www.proteinatlas.org/ENSG00000066279-ASPM/pathology/tissue/breast$+$cancer#img). (C) Protein levels of CDC20 in normal tissues (https://www.proteinatlas.org/ENSG00000117399-CDC20/tissue/breast#img). (D) Protein levels of CDC20 in tumor tissues (https://www.proteinatlas.org/ENSG00000117399-CDC20/pathology/tissue/breast$+$cancer#img). (E) Protein levels of TTK in normal tissues (https://www.proteinatlas.org/ENSG00000112742-TTK/tissue/breast#img). (F) Protein levels of TTK in tumor tissues (https://www.proteinatlas.org/ENSG00000112742-TTK/pathology/tissue/breast$+$cancer#img).

Figure 9

Experimental validation of ASPM, CDC20 and TTK. (A) Relative expression of ASPM, CDC20, and TTK in breast cancer tissues and paracancerous tissues. (B) Cell Counting Kit-8 (CCK8) assay. (C) Clone formation assay. ^***P < 0.001. Student's t-tests were used to evaluate the statistical significance of differences.

Figure 10

Gene set enrichment analysis. (A) The top 5 enriched pathways in samples with ASPM high expression. (B) The top 5 enriched pathways in samples with CDC20 high expression. (C) The top 5 enriched pathways in samples with TTK high expression.

Figure 11

Correlations among hub genes. (A) Correlation between CDC20 and TTK in GSE25055. (B) Correlation between CDC20 and TTK in GSE42568. (C) Correlation between ASPM and CDC20 in GSE25055. (D) Correlation between ASPM and CDC20 in GSE42568. (E) Correlation between ASPM and TTK in GSE25055. (F) Correlation between ASPM and TTK in GSE42568.