Introduction: Breast cancer is a molecularly heterogeneous disease. The existing molecular classifications provide a good introduction of the molecular heterogeneity. However more efforts are necessary to characterize molecularly distinct tumor subgroups that are likely responders to novel targeted therapy interventions. One of the classification strategies is to characterize robust classes based on activity of specific biological pathways and networks. This thesis describes a classification based on activity status of p53, ER and VEGF signaling in breast cancer.
Methods: In paper I, canonical sequences from the proximal promoter of the genes that distinguish the molecular portraits are studied to identify the significantly overrepresented potential TFBS motifs for each molecular class. Subtype-specific networks based on previously reported or predicted interactions between subtypespecific genes and corresponding transcription factors. In paper II, breast cancer expression profiles were analyzed for inferring differentially activated pathways and differentially expressed genes by p53 gene mutation status using geneset-based and individual gene search methods. Genes are evaluated for prognostic significance. Paper III compares miRNA and mRNA expression profiles from the same sample set by VEGF mRNA expression status and for each differentially expressed miRNA, class-specific potential targets are the mRNAs having target site (based on the target database) and are differentially expressed as well as class-specific anti-correlated to their potential regulator miRNA.
Results: We identified the significantly overrepresented potential TFBS motifs by subtype and showed positive correlation between the subtype-specific mRNA expressions of some of their corresponding TF genes and degree of TFBS overrepresentation. The network analysis showed p53 as a topological hub that has interactions with subtype-specific genes thus explaining core functional significance of p53 signaling (Paper I). We also identified about 40 pathways differentially activated by the p53 mutation status. Besides VEGF expression was shown to predict survival after controlling for p53 mutation status and subtype. In ER+/PR+ patients, effect of VEGF was found significant but not in ER–/PR– patients (Paper II). MiRNA profiles of VEGF upregulated group showed upregulation of miR-590-5p, miR-18a/18b/19a cluster, miR- 9/9*, miR-135b, and downregulation of miR-149, miR-342-3p/5p, miR- 449a. The anti-correlated targets of upregulated miRNAs were enriched for angiogenesis pathway, vasculature development, TGF-ß signaling and focal adhesion. Anti-correlated targets of downregulated miRNAs in VEGFA+ group were associated with EGFR pathway, positive regulation of DNA binding and nucleolus (Paper III). This work implicates experimental validation.
List of papers. Paper III is removed from the thesis due to publisher restrictions.
Paper I: Overrepresentation of transcription factor families in the genesets underlying breast cancer subtypes. Joshi H, Nord S, Frigessi A, Børresen-Dale AL, Kristensen V. BMC Genomics. 2012 May 22;13:199. doi:10.1186/1471-2164-13-199 2012 Joshi et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License
Paper II: Potential tumorigenic programs associated with TP53 mutation status reveal role of VEGF pathway. Joshi H, Bhanot G, Børresen-Dale AL, Kristensen V. Br J Cancer. 2012 Nov 6;107(10):1722-8. doi:10.1038/bjc.2012.461 This work is licensed under the Creative Commons Attribution-NonCommercial-Share Alike 3.0 Unported License.
Paper III: Implications of VEGFA upregulation on microRNA-mRNA Modules in Breast Cancers. Joshi H, Børresen-Dale AL, Kristensen V. Manuscript