Bioinformatic Analysis And Functional Studies Of Breast Cancer Metastasis Related Genes

Background&ObjectiveBreast cancer is the most common malignant disease in women. In these patients, it is not the primary tumor, but its metastases at distant sites that have a closely related to their treatment and prognosis, and the main cause of death. At present, the prognostic indicators of breast cancer is primarily based on clinical stage, histological type and lymph node metastasis, but which are difficult to assess the metastatic potential and condition accurately. Approximately10%-15%of patients with breast cancer have an aggressive disease and develop distant metastases within3years or more after the initial detection of the primary tumor. Therefore, patients with breast cancer are therefore at risk of experiencing metastasis for their entire lifetime. Metastasis and recurrence is the main obstacle for the living of the patients with breast cancer. To study the molecular target about invasion and metastasis of breast cancer cell is significance for preventing the recurrence of breast cancer and improving the therapeutic effect after surgical resection.Metastasis of breast cancer is a complex process which multi-gene and multi-factor interact and influence mutually like other malignancies, and is also a trouble problem in tumor prevention and therapy. Only when metastasis successfully blocked, the complete cure of malignant tumor can be reached. In the course of tumor development some tumor cells acquire invasive and metastatic potency which were induced by genetic alteration of tumor cells. It is a hot spot of research to search for the metastasis associated genes and reveal the molecular mechanisms of metastasis. The microarray technology developed in recent years shows its superiority in the analysis of diseases.The gene chip (DNA microarray) is a major technique in bioscience field in the21st century, which has the outstanding of high-throughput and rapid detection. Gene expression profiling chip has the superiority on the study of gene expression profiling mode of cells, which can be used to study the change of gene expression profiling in every stage of tumor cell development, understand the gene expression mode during the occurrence and development of tumors. Thus, gene expression profiling chip has widespread application in tumorigenesis, early diagnosis, tumor genotyping, directing therapy and prognosis evaluation, etc. With the rapid development and widespread application of gene expression profiling chip, redundant and complicate biological data was arised. How to analyze tens of thousands gene hybridization information on the gene-chip to reveal the life characteristics and regular pattern, which has become the major "bottleneck" restricting the further development of gene expression profiling chip. The analysis of tumor expression profiling chip is useful to clarify the mechanism of tumorigenesis, the diagnosis and therapy of tumor. Bioinformatics is a crossing science that conjunct computing-technology, statistics, mathematics, genetics, molecular biology, medical science, informatics, bio-physics and chemistry, which can be applied for the data-mining and explore the disease from the molecular level based on the bio-chips, sequences aligning, statistics methods, visualization, biological clustering, pathways analysis and promoter prediction. It is an important approach for functional genomics to obtain tumor-related genes and regulatory mechanism using gene-chip technologies and bioinformatics, which can integrate the gene expression network of tumors from the genome level, provide new insights and open up a new way to explore the pathogenesis of tumors.Three gene expression profiling datesets related to the breast cancer metastasis (GSE2034, GSE2603and GSE12276), which were downloaded from gene expression omnibus (GEO), were used in this study. GeneSpring software was used to identify the differential expressed genes between the primary tumors and metastatic tumors of breast cancer, and bioinformatics tools and literature mining were used to analyze the relationship of the differential genes, which found a new breast cancer metastasis-associated gene COL1A1, and then its function was studied by molecular biological experimental methods, which provide a new idea for the pathogenesis metastasis of breast cancer, and lay the foundation for the molecular diagnosis and individualized treatment of metastatic breast cancer.Methods1. Inputing the gene expression profiling datasets of GSE2034(172Primary Breast Cancer samples VS98metastatic breast cancer samples), GSE2603(47primary breast cancer samples VS24metastatic breast cancer samples) and GSE12276(21primary breast cancer samples VS metastatic breast cancer samples) into GeneSpring GX11.5software to screen differential genes between primary tumor and metastatic tumor of breast cancer. Then, we cross-compared the three differential expression groups to obtain co-overexpressed and co-underexpressed genes. Biological methods including PANTER, GATHER, GSEA, STRING and pSTIING were used to analyze these differential genes and construct the network of differential target genes. And then a new metastasis-associated gene was identified by iHOP literature mining tool.2. Based on the human COL1A1mRNA gene sequencing,2sequences targeting COL1A1gene and1scramble sequence were designed and synthesized. pSilencer2.1-U6/neo plasmid was used to construct the specific RNA interference vector (pshRNA-COL1A1-1, pshRNA-COL1A1-2) targeted COL1A1gene mRNA and pshRNA-scramble contained a scramble sequence. The constructed siRNA expression vectors of positive clones were validated by enzyme digestion and DNA sequencing.3. Transfected MDA-MB-231cells with the reconstructed plasmids by cationic lipsome. Screen positive stable transfection cell clones. Detected the change of COL1A1mRNA and protein expression level with RT-PCR and Western blot respectively. The change of the cell cycle and cell apoptosis were Determined by flow cytometry, cell morphological observation by Hoechst33258, cell proliferation activity by MTT chromatometry and the assay of colony formation in plate, cell adhesion by plate adhesion model and the mobility of the cells by Transwell assay.Results1. Identification and bioinformatic analysis of breast cancer metastasis related genes147differentially expressed genes were identified by the GeneSpring GX11.5software, of which93genes were over-expressed,54genes underexpressed. Pathway analysis and functional annotation of these differentially expressed genes by the GATHER, PANTHER bioinformatics tools and GSEA methods found that these genes associated with the following biological pathways:cell cycle and proliferation, cell adhesion, cell migration, angiogenesis and signal transduction. To further understand the interaction networks among147differentially expressed genes, these proteins interactions of which encoded by these differentially expressed genes were analyzed by STRING online tools and found that the interaction among the proteins were mainly concentrated in the14proteins, such as TNC, SPARC, PTGS2, TGFB2NGF CTGF, FN1, SMAD1, MEF2C, CSF1, CXCR4, MMP1, MMP2, MMP3, COL1A1and ANGPL4, and then the pSTIING bioinformatics tool was used to construct the network related to transcript of these14differential target genes.9(CXCR4, MMP1, MMP2, MMP3, CTGF, COL1A1, MEF2C, PTGS2and SPARC) of which still played important role in a more complex network map position, and finally the literature mining for the9important junction gene was done by iHOP online software to find that COL1A1may be a new metastasis related gene of breast cancer.2. Construction and identification of specific pshRNA-COL1A1expression vector.1) The constructed siRNA expression vectors of positive clones were confirmed by enzyme digestion and DNA sequencing. The results showed that inserted fragments of positive clones were consistent with what expected.2) The constructed siRNA expression vectors were transfected into MDA-MB-231cells by cationic lipsome. The cell clones of positive stable transfection were screened. The results of RT-PCR showed that the mRNA expression level of COL1A1decreased significantly in pshRNA-COL1A1-1and pshRNA-COL1A1-2groups compared with that in control group(P<0.05), The inhibitory rates of COL1A1mRNA level were44.407±3.900%and63.050±3.131%, respectively, but there was no significant defference between pshRNA-scramble and control group.3) The result of Western blot showed that the expression level of COL1A1protein decreased significantly in pshRNA-COL1A1-1and pshRNA-COL1A1-2groups compared with that in control group (P<0.05), The inhibitory rates of COL1A1mRNA level were45.495±2.711%and66.984±2.081%, respectively.3. The influences on cell biology of breast cancer cell line MDA-MB-231by COL1A1silence.1) A significantly time-dependent inhibitory proliferation was found in pshRNA-COL1A1-2cells compared with pshRNA-scramble and control by in vitro MTT assay(F=31.597, P=0.000); Flow cytometry results showed significant difference among the three cells in G0/G1phase and S phase change (P<0.05). COL1A1silence effected cell cycle and prevented GO/G1phase cells into the S phase, and showed significant apoptosis in MDA-MB-231cells of pshRNA-COL1A1-2group compared with that of control group(P=0.000), there was no significant difference between pshRNA-scramble group and control group. In addition, pshRNA-COL1A1-2cells had a significant impaired ability to form colonies in plates, as compared with pshRNA-scramble and control cell (F=601.181, P=0.003). The results indicate that COL1A1silence suppressed proliferation of MDA-MB-231cells.2) pshRNA-COL1A1-2cells had a significantly reduced ability of adhesion, migration and invasion as compared with pshRNA-scramble and control cells detected by Plate adhesion mode and Transwell assay in vitro(F=37.376, P=0.007),(F=427.574, P<0.001) and (F=79.625, P=0.011), respectively.ConclusionTo conducted in-depth bioinformatics analysis for the differentially expressed genes with breast cancer metastasis by the use of gene expression data analysis tools, bioinformatics tools and literature mining tool and screen nine breast cancer metastasis-related genes including the COL1A1gene with breast cancer metastasis reported for the first time. Successfully designed and constructed specific RNA interfere plasmid targeting COL1A1gene mRNA. The MDA-MB-231was cultured and transfected stably with specific-COL1Al siRNA interference plasmid. In vitro, specific-COL1A1siRNA could downregulate COL1A1mRNA and protein expression in MDA-MB-231cells, make the cell cycle arrest at G0/G1phase, significantly suppress the function of growth, proliferation, adhesion, migration and invasiveness, and promote apoptosis of the MDA-MB-231cells. Which provided some evidences for RNAi technology in preventing and treating breast cancer.