Control of gene expression by dynamic metabolic oscillations

Summary Emerging evidence shows that changes in cellular metabolism can induce broad shifts in gene expression, but the mechanisms underlying this connection are not fully understood. Previous work has not examined the impact of temporal dynamics on metabolism-induced gene expression. Recent work in systems biology has shown that oscillations in upstream inputs can be filtered by gene expression machinery to modulate gene expression, through a process termed dynamic filtering. Additionally, we have recently shown that cellular metabolic status fluctuates rapidly in response to various forms of metabolic stress. These cycles drive asynchronous oscillating activity of transcription factors including FOXO3, a key regulator of stress genes that plays a role in aging, and TFEB, a central regulator of lysosome and autophagy genes. We therefore hypothesize that oscillations in metabolic state drive gene expression programs that are distinct from those under static unstressed conditions. We propose that dynamics-sensitive gene expression programs can influence cell fate decisions such as differentiation, cell growth, senescence, inflammation, and drug sensitivity. In this project, we will investigate how metabolic oscillations control the expression of TFEB and FOXO3 target genes. We will use live-cell reporters, inducible expression constructs, and other methods to measure key kinetic parameters in the transcription and translation of target genes. To identify broader gene expression programs modulated by metabolic dynamics, we will use mathematical modeling in combination with transcriptome-level profiling. Functional assays will be used to test how dynamically sensitive gene expression programs alter cell fates. We expect our study to establish an important unexplored mechanism that explains how short-term regulation of cellular metabolic status influences chronic diseases including cancer, diabetes, and aging. Our results will address the outstanding question of how pharmacological metabolic inhibitors such as metformin provide benefits in cancer and aging. The models generated will establish a new approach to evaluate candidate pharmacological compounds.