Systems-Level Principles and Mechanisms Underlying Cellular Adaptation

About This Grant

Summary The goal of our research is to discover the central principles that govern cellular adaptation. We aim to understand how cells achieve adaptive gene-expression states, both during short-term physiological adaptation and long-term adaptive evolution. We investigate these phenomena on a systems-level, often necessitating observations, perturbations, or analyses that are beyond the scale and precision of existing methods. Thus, our laboratory also develops new enabling technologies and computational methods. In this R35 application, we seek support for three NIGMS-related projects: (1) Cellular adaptation by stochastic tuning of gene expression. We have discovered a powerful new mechanism, that we call stochastic tuning, by which eukaryotic cells adapt to extreme or novel challenges. During stochastic tuning, cells utilize transcriptional noise to randomly change the expression of individual genes, and to actively reinforce those changes that improve the overall health of the cell. Stochastic tuning therefore enables cells to prospectively explore novel gene expression states that enable adaption to challenges in real time—including conditions never previously encountered—thereby bypassing the need for pre-determined hardwired regulatory programs. We have compelling new evidence that stochastic tuning is the key underlying mechanism for non-mutational cancer chemotherapy resistance, recognized as a major barrier to effective cancer therapies. We are utilizing CRISPR-interference and largescale reporter assays to define the critical protein and DNA effectors of stochastic tuning in yeast and to mechanistically determine their roles using chemical/genetic/optogenetic perturbations of single cells in well-controlled microfluidic experiments. (2) Genetic basis of microbial habitat adaptations. We have developed a versatile computational framework to conduct genotype-habitat association at the tree-of-life scale, enabling discovery of genes that underlie microbial colonization of specific habitats. By applying this analysis to the gut microbiome, we have discovered many highly conserved factors that strongly contribute to gut colonization. We are using functional genomics technologies to efficiently determine the molecular mechanisms by which these factors enable gut colonization. In addition, we are developing state-of-the-art deep learning and protein language models to improve the sensitivity/specificity of genotype-habitat association, enabling large-scale microbial engineering for diverse biomedical applications. (3) Global mapping of all-against-all molecular interactions in a single tube. We have recently developed a powerful technology for coupling in vivo expressed proteins to their encoding messenger RNAs, enabling a diverse array of proteomic assays to be performed by using DNA- sequencing as a readout. We propose to develop this platform to enable routine comprehensive all-against-all protein-protein and protein-DNA interaction studies on the timescale of days. This technology promises to transform our ability to rapidly map molecular network interactions under dynamic physiological conditions, an essential capability in the era of AI-enabled biology.