Uncovering Coordinated Regulation Of Transcription Factors By Micrornas Using Integrated Network Models

MicroRNAs (miRNAs) are small non-coding RNAs that play a significant role in development and cancer but induce only moderate repression of direct messenger RNA (mRNA) targets, suggesting that they coordinate with other modes of cellular regulation to affect cellular phenotype. In this work we exhaustively profile the transcriptional and post-transcriptional regulatory changes between an isogenic pair of murine fibroblast cell lines with and without Dicer, an enzyme required for miRNA processing. Through quantitative analysis of diverse high-throughput datasets collected from both cell lines, we were able to decouple transcriptional from post-transcriptional changes reflected in mRNA expression changes between cells with and without miRNAs. We present three major findings. First, we find that direct miRNA-mRNA interactions have a limited impact on gene expression changes upon Dicer deletion when compared to changes at transcription. We then introduce an integrative graphical network approach to identify specific transcription factors that explain the observed changes in gene expression upon loss of Dicer. Validation of this computational model via transcription factor over-expression reveals a subset of the miRNA-mediated transcriptional program that is activated upon Dicer loss, confirming that transcriptional networks amplify the effects of miRNAs. Lastly, we use this network to examine coordination between transcriptional and post-transcriptional regulation. We identify numerous coherent and in-coherent feed-forward loops, network motifs that have been alluded to but only minimally studied in the context of microRNAs and transcription factors, and find that genes regulated within feed-forward loops by microRNAs and transcription factors exhibit distinct properties in development. In summary, our work illustrates how transcriptional networks amplify the effects of microRNAs. We show coordinated regulation of transcription factors by microRNAs that suggests that most gene expression changes attributed to microRNAs in diseases such as cancer are due to changes in transcription rather than microRNA-mediated degradataion. As such, this work has the potential to reframe future studies of microRNAs in the context of cancer by shifting the focus to understanding the impact of microRNAs on transcription factors. Citation Format: Sara JC Gosline, Allan M. Gurtan, Courtney K. JnBaptiste, Andrew Bosson, Pamela Milani, Simona Dalin, Bryan Matthews, Yoon Sing Yap, Phillip A. Sharp, Ernest Fraenkel. Uncovering coordinated regulation of transcription factors by microRNAs using integrated network models. [abstract]. In: Proceedings of the AACR Special Conference on Computational and Systems Biology of Cancer; Feb 8-11 2015; San Francisco, CA. Philadelphia (PA): AACR; Cancer Res 2015;75(22 Suppl 2):Abstract nr B2-45.