Systematic analysis of the expression and prognosis relevance of FBXO family reveals the significance of FBXO1 in human breast cancer

Abstract

Background: Breast cancer (BC) remains a prevalent and common form of cancer with high heterogeneity. Making efforts to explore novel molecular biomarkers and serve as potential disease indicators, which is essential to effectively enhance the prognosis and individualized treatment of BC. FBXO proteins act as the core component of E3 ubiquitin ligase, which play essential regulators roles in multiple cellular processes. Recently, research has indicated that FBXOs also play significant roles in cancer development. However, the molecular functions of these family members in BC have not been fully elucidated.

Methods: In this research, we investigated the expression data, survival relevance and mutation situation of 10 FBXO members (FBXO1, 2, 5, 6, 16, 17, 22, 28, 31 and 45) in patients with BC from the Oncomine, GEPIA, HPA, Kaplan-Meier Plotter, UALCAN and cBioPortal databases. The high transcriptional levels of FBXO1 in different subtypes of BC were verified by immunohistochemical staining and the specific mutations of FBXO1 were obtained from COSMIC database. Top 10 genes with the highest correlation to FBXO1 were identified through cBioPortal and COXPRESdb tools. Additionally, functional enrichment analysis, PPI network and survival relevance of FBXO1 and co-expressed genes in BC were obtained from DAVID, STRING, UCSC Xena, GEPIA, bc-GenExMiner and Kaplan-Meier Plotter databases. FBXO1 siRNAs were transfected into MCF-7 and MDA-MB-231 cell lines. Expression of FBXO1 in BC cell lines was detected by western-blot and RT-qPCR. Cell proliferation was detected by using CCK-8 kit and colony formation assay. Cell migration was detected by wound-healing and transwell migration assay.

Results: We found that FBXO2, FBXO6, FBXO16 and FBXO17 were potential favorable prognostic factors for BC. FBXO1, FBXO5, FBXO22, FBXO28, FBXO31 and FBXO45 may be the independent poor prognostic factors for BC. All of them were correlated to clinicopathological staging. Moreover, knockdown of FBXO1 in MCF7 and MDA-MB-231 cell lines resulted in decreased cell proliferation and migration in vitro. We identified that FBXO1 was an excellent molecular biomarker and therapeutic target for different molecular typing of BC.

Conclusion: This study implies that FBXO1, FBXO2, FBXO5, FBXO6, FBXO16, FBXO17, FBXO22, FBXO28, FBXO31 and FBXO45 genes are potential clinical targets and prognostic biomarkers for patients with different molecular typing of BC. In addition, the overexpression of FBXO1 is always found in breast cancer and predicts disadvantageous prognosis, implicating it could as an appealing therapeutic target for breast cancer patients.

Keywords: Bioinformatics analysis; Biomarkers; Breast cancer; F-box protein; Prognosis.

Fig. 1

The transcription levels of 10 FBXO members in Breast Cancer. a The Expression of FBXOs in BC in Oncomine Database. Red, over-expression; Blue, down-regulated expression. b The scatter diagram of Expression of FBXOs in BC in GEPIA Database. c The box plot of Expression of FBXOs in BC in GEPIA Database. T, Tumor tissues; N, Normal tissues; * P-value was set at 0.01

Fig. 2

The Transcriptional Levels of FBXOs in Different Subtypes of BC in GEPIA Database. T, Tumor tissues; N, Normal tissues; * P-value was set at 0.01

Fig. 3

Correlation between FBXOs Expression and Tumor Stage of BC Patients in UALCAN Database. *P < 0.05; **P < 0.01; *** P < 0.001; ****P < 0.0001

Fig. 4

The Comparison of Protein Expression of FBXOs between BC and Normal Tissues from Human Protein Atlas (HPA). FBXO1, FBXO5, FBXO6, FBXO16, FBXO45 proteins were highly expressed in BC than in the normal tissues. The expression differences of FBXO2 and FBXO31 were not obvious between tumor and normal tissues

Fig. 5

The mutation information of FBXO genes in human BC (cBioPortal). a Alteration frequency of 10 FBXO genes at the overall and individual levels. b The bar graph of gene alteration frequency of FBXO genes. c Schematic representation of gene mutation sites of FBXOs on the coding strand

Fig. 6

The prognostic values of FXBO family members in BC patients. The survival curves comparing BC patients with high (red) and low (black) FBXO expression levels were plotted using the Kaplan–Meier Plotter. DFS, disease-free survival rate; OS, the overall survival rate; DMFS, distance metastasis free survival; PPS, post progression survival; The threshold P-value is less than 0.05

Fig. 7

Comprehensive Bioinformatics Analysis of FBXO1 and Co-expressed Genes in BC. a High expression of FBXO1 protein in different subtypes of BC verified by immunohistochemistry. Scale bar = 50 μm, T, tumor tissues; N, normal tissues. b The percentages of mutation types of FBXO1 in BC were indicated in a pie chart generated from Catalogue of Somatic Mutations in COSMIC database. c, d The top 150 genes positively associated with FBXO1 transcript level based on the cBioPortal and COXPRESdb databases in BC. The tables listed the top 20 genes. e Venn diagram represented the intersection of top positively corrected genes between the cBioPortal and COXPRESdb databases. f The bubble diagram showed the functions of FBXO1 and 108 genes significantly associated with FBXO1 alterations, which were predicted by the analysis of Kyoto Encyclopedia of Genes and Genomes (KEGG) by STRING tools. g Gene Ontology (GO) enrichment analysis predicted the functional roles of FBXO1 and 108 co-expressed genes based on three aspects, including biological processes (BP), cellular components (CC) and molecular function (MF)

Fig. 7

Comprehensive Bioinformatics Analysis of FBXO1 and Co-expressed Genes in BC. a High expression of FBXO1 protein in different subtypes of BC verified by immunohistochemistry. Scale bar = 50 μm, T, tumor tissues; N, normal tissues. b The percentages of mutation types of FBXO1 in BC were indicated in a pie chart generated from Catalogue of Somatic Mutations in COSMIC database. c, d The top 150 genes positively associated with FBXO1 transcript level based on the cBioPortal and COXPRESdb databases in BC. The tables listed the top 20 genes. e Venn diagram represented the intersection of top positively corrected genes between the cBioPortal and COXPRESdb databases. f The bubble diagram showed the functions of FBXO1 and 108 genes significantly associated with FBXO1 alterations, which were predicted by the analysis of Kyoto Encyclopedia of Genes and Genomes (KEGG) by STRING tools. g Gene Ontology (GO) enrichment analysis predicted the functional roles of FBXO1 and 108 co-expressed genes based on three aspects, including biological processes (BP), cellular components (CC) and molecular function (MF)

Fig. 8

The knockdown of FBXO1 attenuates the proliferation and migration of breast cancer cells in vitro. a Upper panel, the expression levels of FBXO1 protein examined by Western blotting in 8 human breast cell lines. Lower panel, bar graphs representing quantification of Western blotting bands. b Determination of relative mRNA expression levels of FBXO1 in control and si‐FBXO1‐transfected MCF7 and MDA-MB-231 breast cancer cell lines by RT-qPCR assay. c Immunoblotting analyses of proteins as indicated in control and si‐FBXO1‐transfected MCF7 and MDA-MB-231 cell lines, bar graphs representing quantification of Western blotting bands. d Diagram of successful transfection of siRNA of FBXO1 labeled by FAM fluorescence dye in MCF7 and MDA-MB-231 cell lines. e The knockdown of FBXO1 attenuates the proliferation of breast cancer cells in vitro. Cell Counting Kit-8 assay showed the relative proliferative capacity of specific MCF7 and MDA-MB-231 cells at 24, 48, and 72 h after seeding in plates. f The knockdown of FBXO1 attenuates the proliferation of breast cancer cells in vitro. Colony-forming assay showed the relative proliferative capacity of specific MCF7 and MDA-MB-231 cells at 48 h after seeding in plates(left) and quantification of the colony areas (right). g The knockdown of FBXO1 attenuates the migration of breast cancer cells in vitro. Transwell migration assay showed representative images of specific MCF7 and MDA-MB-231 cells (left) and quantification of the cell numbers (right). h The knockdown of FBXO1 attenuates the migration of breast cancer cells in vitro. Wound‐healing assay for MCF7 and MDA-MB-231 and wound closure was monitored at 0, 24, and 48 h. Data in bar graphs are the means ± SD of three independent experiments. **P < 0.01; ***P < 0 .001 by Student t test. siRNA, small interfering RNA; RT-qPCR, Real Time Quantitative Polymerase Chain Reaction

Fig. 9

Protein–protein Interaction (PPI) Network and Correlative Analysis of FBXO1. a PPI network of FBXO1 and 108 co-expressed genes. The most significant modules and hub genes of the PPI network were analysed by Cytoscape software, which were marked in yellow. b The hub-genes were identified using cytoHubba tool kits in Cytoscape. c The PPI network of hub-genes were identified using STRING database. d The biological process analysis of hub-genes was performed using the BiNGO plug-in. P < 0.05 was considered to be a statistically significant difference. e The hierarchical clustering of hub-genes was constructed using UCSC online database. f The heat map of correlation between FBXO1 and hub-genes in BC patients analysed by bc-GenExMiner v4.0. Different color represents the percentage of correlationship. 1 stands for complete positive correlation and -1 stands for complete nagetive correlation

Fig. 10

Correlative and Survial Analysis of Hub-genes. a Correlation between 10 hub-genes and FBXO1 mRNA expression determined by GEPIA database. b RFS survival analyses of hub-genes in BC. The results based on the KM plotter database indicated that all hub-genes were associated with poor prognosis in BC