Protein Biosensors with Customized Properties for Live-Cell Imaging

DESCRIPTION (provided by applicant): Live-cell fluorescence microscopy provides spatially resolved kinetic information about dynamic processes as they occur in cells, and therefore it has unique potential for the quantitative measurement of signal transduction and other intracellular processes. This approach has not yet lived up to its full potential, however, because it is currently limited by the availability of well-characterized biosensor molecules that can be either genetically encoded or microinjected into cells and which recognize intracellular target molecules of interest. What is not appreciated by most practitioners of live-cell imaging is that the binding affinity of a suitable biosensor for its target(s) must lie within a relatively narrow range, and the reliance on modular binding domains derived from naturally occurring proteins presents a major constraint in that regard. Herein, we propose a collaborative effort to develop a novel and general approach enabling de novo design of intracellular biosensors with prescribed properties for live-cell imaging. To do this, it is clear that we must move away from binding domains derived from naturally occurring proteins. Instead, our starting point is an ensemble of 7 different proteins from hyperthermophilic archaea and bacteria (which thrive at extreme temperatures) as templates, or scaffolds, for engineering biosensors. These scaffolds have several favorable properties; they are low in molecular weight (~100 amino acids or less), lack disulfide bonds, and are functionally inert in mammalian cells. Proteins with desired binding specificity will be screened from a large (> 108) combinatorial collection of mutant proteins that we have successfully generated through randomization of surface-accessible residues on the protein scaffolds. Because the protein engineering strategy uses multiple scaffolds, we refer to the repertoire of variants as a super library, which possesses greater theoretical diversity than a library of the same size derived from any single scaffold. We propose to screen the super library for biosensors recognizing specific phosphorylation sites and characterize their binding affinities (Aim 2). Subsequently, we will validate the identified biosensors by characterizing thei translocation in cells expressing wild-type or mutant Epidermal Growth Factor Receptor (EGFR) (Aim 2). Finally, we propose a pilot study to assess the kinetics of site-specific EGFR phosphorylation in human mammary epithelial cells (Aim 3).