Abstract

Using Image-omics to Understand the Relationship between Cell Shape and Transcription in Cancer Cells

Chris Bakal, Dynamical Cell Systems Team Leader, Institute of Cancer Research

In order to execute diverse cellular behaviors, including proliferation, migration, and differentiation, as well as perform tissue-specific tasks, cells must adopt different shapes. Cell shape is the emergent behavior of a system capable of integrating mechanical, geometric, and soluble cues. The system underpinning cell morphogenesis is largely comprised of biochemical networks that act to sense these cues, and convert them into chemical signals that regulate transcription factor activation. By controlling gene expression, transcription factors can alter shape, and generate feedback on the network. Thus through the actions of regulatory networks cell shape orchestrates transcription, and in turn transcription influences cell shape. Because cell morphogenesis is vital to development and homeostasis, and often dysregulated during the progression of many diseases including cancer, obtaining a quantitative understanding of the regulatory networks that couple cell shape to transcription is warranted, and will open up therapeutic avenues. Using compuational approaches to analyze single cell shape ‘signatures’ generated by high-content imaging of a panel of breast cancer lines, we discovered that the activity of the pro-inflammatory transcription factor NF-kappaB is regulated by cell shape (Sero et al., Molecular Systems Biology 2015). Recently using mathematical models in tandem with functional genomics we identifying a new mechanisms coupling cell shape to the dynamics of the YAP transcription factor (Sero et al., Cell Systems 2017). To gain both systems-level insights into the relationship between cell shape and transcription we have developed an image-omics approach to integrate quantitative cell imaging data, gene expression, and protein–protein interaction data to systematically describe a “shape-gene network”. The actions of this network converge on NF-?B, and support the idea that NF-?B is responsive to mechanical stimuli. Critically, these networks have predictive value regarding tumor grade and patient outcomes. Our work highlights the importance of integrating phenotypic data at the molecular level with those at the cellular and tissue levels to better understand breast cancer oncogenesis (Sailem et al., Genome Reserarch 2017).