Computational design of a protein-based coenzyme a biosensor

Official URL: https://risc01.sabanciuniv.edu/record=b2766928_(Table of contents)

Cell environment comprises many small molecules involved in metabolic processes along with proteins to carry out its routine work. Since changes in the intracellular concentrations of these molecules are indicative of cellular abnormalities, it is crucial to monitor and measure these molecules in situ. With this regard, genetically encoded fluorescent biosensors (GEFBs) have come into prominence as they enable real-time measurement of dynamic events in cell by using the intrinsic fluorescence property of proteins. The GEFBs, basically consisting of an analyte-sensing domain and fluorescent protein (FP), are based on the principle that the conformational change in the protein upon binding of the analyte of interest triggers the chromophore microenvironment of the FP, thus giving rise to a detectable change in the fluorescence yield. Developing GEFBs requires a long trial-and-error process. However, in theory, it is possible to rationalize these design steps using computational methods and to make effective interventions to the design at the molecular level. With this motivation, in the scope of the hypothesis that the GEFBs design problem can only be optimized with a holistic understanding of the structure-dynamics of proteins, we developed a computational workflow that can provide an initial design idea for GEFB construction. To this end, we selected coenzyme a (CoA) molecule as a target analyte, and developed possible design models for single circularly permuted FP-based CoA GEFBs. Our pipeline not only provides design models for CoA biosensor construction, but also paves the way for the computational design of FP-based biosensors for any generic analyte.