Linearity in Cell Signaling Pathways

Citation

Nunns, Harry James Rogan (2019) Linearity in Cell Signaling Pathways. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/WP0J-E945. https://resolver.caltech.edu/CaltechTHESIS:01222019-133152032

Abstract

Accurate cellular communication is of paramount importance for the development, growth, and maintenance of multi-cellular organisms. Communication between cells is carried out by a highly conserved set of signaling pathways, whose dysregulation can lead to many diseases. The molecular details of these signaling pathways are now well-characterized, allowing researchers to investigate the emergent properties that arise from the complex signaling networks. These properties often arise from counter-intuitive or paradoxical mechanisms, meaning that systems-level analysis is necessary. Importantly, mathematical models have been constructed for many pathways that capture measured reaction rates and protein levels. These mathematical models successfully recapitulate dynamic responses of each pathway. Here, I investigated the input-output response of the Wnt, MAPK/ERK, and Tgfβ pathways using analytical and numerical treatment of mathematical models. Using this approach, I found that the distinct architectures of the three signaling pathways lead to a convergent behavior, linear input-output response. Specifically, mathematical analysis reveals that a futile cycle in the Wnt pathway, a kinase cascade coupled to feedback in the ERK pathway, and nucleocytoplasmic shuttling in the Tgfβ pathways all yield linear signal transmission. I then verified this finding experimentally in the Wnt and ERK pathways. For the Wnt pathway, direct measurements of the input-output response reveal that β-catenin is linear with respect to Wnt co-receptor LRP5/6 activity up until receptor saturation. For the ERK pathway, direct measurements indicate a linear relationship between phosphorylated ERK1/2 and the concentration of EGF ligand, up until saturation of ERK1/2. Finally, mathematical modeling reveals that linear response in the Wnt pathway, in conjunction with a recently identified cis-regulatory motif, is sufficient to explain gene expression buffering to perturbations. Therefore, this thesis demonstrates how linearity emerges across three dissimilar architectures, and introduces a novel benefit for linear signal transmission in biology.

Thesis Files