Fluorescent protein biosensors have broad applications as probes of biological function. Protein phosphorylation and signal transduction cascades control cellular function, and often are changed in cancer to permit uncontrolled cell growth and invasion. Encoded fluorescent probes of protein phosphorylation have been applied to understand numerous kinases, but most have substantial limitations due to the small fluorescence changes upon phosphorylation (typically 5-10% change in fluorescence) and the large size of the constructs (typically greater than 60 kDa). We will develop a novel approach to the fluorescent detection of protein kinase activity based on the design of a phosphorylation-dependent version of the proto-fluorescent protein UnaG. UnaG fluorescence depends on the binding of the ubiquitous metabolite bilirubin. We will reengineer UnaG so that it binds bilirubin poorly, and thus is non-fluorescent, when it is non-phosphorylated. In contrast, upon phosphorylation, UnaG will bind bilirubin tightly and become fluorescent. Because the fluorescence of UnaG is reversible, in contrast to the fluorescence of GFP and GFP derivatives, the reversible binding of bilirubin can form the basis of phosphorylation-dependent fluorescence. The small size of UnaG allows UnaG-based protein kinase sensors to be employed as encoded kinase-responsive protein tags, allowing localization of the sensors onto proteins of interest to more effectively interrogate intracellular signaling. This approach also is expected to result in much larger fluorescence changes upon phosphorylation than existing FRET-based encoded sensors, with the small size allowing more rapid development of fluorescent protein kinase biosensors. The basis of the design is that UnaG fluorescence derives from the rigid binding of the proto-fluorescent metabolite bilirubin. Thus, mutations which reduce binding or increase dynamics are expected to prevent or reduce fluorescence. The basis of the designs herein is the observation that serine/threonine phosphorylation induces a large disorder-to-order transition. Residues from UnaG will be changed to threonine or serine, with an expected reduction in UnaG fluorescence due to increased dynamics in bilirubin binding. Specific phosphorylation will lead to an increase in order and the restoration of native UnaG fluorescence. The optimized sites of phosphorylation-dependent fluorescence will be identified and applied to incorporate recognition sequences for multiple protein kinases. The phosphorylation-dependent UnaG protein will be applied to examine protein kinase activity of multiple protein kinases in solution and in cells. A ratiometric UnaG-based protein kinase biosensor will also be developed. This work will provide a broad platform for phosphorylation-dependent fluorescent detection of protein kinase activity.
The development of effective treatments for diseases depends on the determination of the specific cellular changes that occur in that disease. One of these significant changes is an increase in the activity of kinase enzymes, which function by adding phosphate groups to proteins (phosphorylation). Changes in phosphorylation have been implicated in the progression of Alzheimer's disease, cancers, and numerous other diseases. Identification of which kinases are acting in a given diseased cell provides a molecular signature of a disease. Fluorescent proteins that detect kinase activities (kinase sensors) are tools to understand the activities of different protein kinases. We will develop a new class of fluorescent protein kinase sensors based on a fluorescent protein from freshwater eels (unagi), the protein UnaG. We will design these proteins to be non-fluorescent when the kinase is not active, but to become fluorescent when the kinase is active, providing a clear signal about the activities of a given kinase. These fluorescent proteins will be new tools to understand the changes occurring in diseases, providing a molecular basis for understanding and treating disease.