Genomic profiles of breast cancer.

摘要

Breast carcinogenesis is caused by molecular genetic changes. These genetic changes ultimately affect the transcriptome. Copy number alteration (CNA) of the genome is a cardinal feature of cancer and plays an important role in tumor progression by altering the gene expression program. These regions of CNA are associated with oncogenes and tumor suppressor genes of known and unknown identity. Characterization of both CNAs and gene expression profiles has been carried out on breast tumor specimens using microarray technology to gain further insight into the progression of this disease.; In the first study, we comprehensively characterized both DNA copy number changes and gene expression profiles of 50 breast cancer cell lines, widely used model systems for the study of breast cancer. We found that the cell lines could be classified into three main subtypes, Luminal, Basal A and Basal B by clustering of gene expression. Overall, this is reminiscent of what is observed in the gene expression analysis of the tumors; however, distinct differences were found. Analysis of the copy number profiles revealed that the cell lines recapitulated the main genetic changes represented in the tumor data set but contained more changes. Our findings inform the selection of appropriate cell line model systems when studying specific processes in breast cancer.; In the second study, we established a strategy to identify candidate oncogenes within an amplicon by investigated a known region found commonly amplified in breast cancer, 17q12 as a proof of principle. Regions of amplification typically contain more than one gene. To identify the driver oncogene and the contribution of each gene to tumorigenesis in 17q12 (STARD3, ERBB2, and GRB7), each gene was knocked down by siRNA respectively and tumor relevant phenotypes were assayed. Knock down of ErbB2 had the greatest impact and therefore was correctly identified as the main oncogene. However, we also observed an effect when STARD3 and GRB7 expression was inhibited. This suggests that multiple genes within an amplicon can contribute to tumorigenesis. STARD3 and GRB7 merit further investigation as potential therapeutic targets.