Transport, assembly, and dynamics in systems of interacting proteins

Many of the functions performed by cells are controlled by networks of interacting proteins. This thesis explores optical techniques for monitoring and perturbing protein networks and measuring protein diffusion. In addition, it discusses the design of simple artificial genetic networks. A useful tool for tagging proteins in diverse organisms is the Green Fluorescent Protein (GFP), which absorbs blue light and emits green light. We report the discovery of a new state of this protein in which it absorbs green light and emits red light. It can be switched to this new state with a pulse of blue light. This effect is useful for marking proteins in a certain region and following their subsequent movement. We use this technique to measure diffusion of proteins inside living bacterial cells. The apparent diffusion coefficient we obtain for GFP is about ten times lower than its value in water. This reflects the crowded conditions in bacterial cytoplasm and places a constraint on the response time of diffusion-based signalling networks. We next turn to systems of motor proteins and microtubules. We show how motor protein forces can be estimated from spiral motions of pinned filaments in motility assays. Then we explore spatially and temporally localized inactivation of specific proteins using Chromophore-Assisted Laser Inactivation (CALI). In this method, dyes are attached to target proteins, allowing them to be inactivated by controlled pulses of light. By inactivating motor proteins with this technique, we can locally disrupt an in vitro assembly process. Finally, we address the problem of constructing an artificial genetic network in order to see whether simple, artificial systems behave as expected from detailed modeling of their components. We analyze the expected behavior of an oscillator built of bacterial transcriptional repressors, and lay the groundwork for its actual construction.