Abstract 4924: Gene network inference models reveal novel key mediators of mammary tumorigenesis

Abstract

Abstract Cancer is a developmental disease, requiring the concerted switching of multiple cellular gene programs. Here we ask whether we can use computational network inference models in a systems biology-driven approach to identify changes in connectivity within the underlying gene regulatory network during cancer progression, as well as novel key mediator genes that are central to altered disease state networks. To investigate this hypothesis, we reverse engineered the gene regulatory networks of mammary glands within normal mice and transgenic C3(1)-SV40Tag mice that undergo mammary gland transformation and tumor progression. RNA was isolated from the mammary glands of transgenic mice at 8, 12, 16 and 20 weeks of age; wild type mice were analyzed at 8 weeks. Whole genome microarray profiles were obtained for each stage. A network inference algorithm was trained on a set of 3000 genome-wide gene expression signatures from public databases. A Mode-of-action Network Inference (MNI) algorithm was used to identify key mediator genes at each stage of the gene regulatory network during different stages of cancer progression in the transgenic mouse mammary gland based on z-score significance. From this initial set of candidates, we focused on key mediator genes classified by gene ontology analysis as transcription factors. To experimentally assess the functional relevance of genes identified by this computational approach, key mediator genes were silenced in vitro in three-dimensional (3D) cultures of mouse mammary tumor epithelial cells derived from the transgenic animals. In contrast to normal mammary epithelial cells, tumor cells failed to form polarized structures containing a lumen in 3D spheroid culture. Importantly, siRNA knockdown of key mediator genes induced these cancer cells to revert to a normalized phenotype in 3D culture. These data demonstrate the feasibility of using genome reverse engineering approaches to identify genes that can potentially induce cancer reversal. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 4924. doi:10.1158/1538-7445.AM2011-4924