Design and Application of Highly Responsive FRET Biosensors for Sugar Detection in Living Yeast

Systems Microbiology Research Center, KRIBB, Daejeon 305-333, Korea; Department of Biotechnology, Yonsei University, Seoul 120-749, Korea

^* To whom correspondence should be addressed. Email: sglee{at}kribb.re.kr.

	Abstract

A protein sensor with a highly-responsive FRET signal for sensing sugars in living yeast was developed by combinatorial engineering of the domain linker and binding protein moiety. Although FRET sensors based on microbial binding proteins have previously been created for visualizing various sugars in vivo, such sensors are limited due to weak signal intensity and narrow dynamic range. In the present study, a FRET-based sensor consisting of CFP-linker₁-maltose binding protein-linker₂-YFP was redesigned as regards the length and composition of the linker moiety, resulting in a ten-fold higher signal intensity. Molecular modeling of the composite linker moieties, including the connecting peptide and terminal regions of the flanking proteins, suggested that an ordered helical structure was preferable for tighter coupling of the conformational change of the binding proteins to the FRET response. When the binding site residue Trp62 of the maltose binding protein was diversified by saturation mutagenesis, the Leu mutant exhibited an increased binding constant (82 µM), accompanied by a further improvement in the signal intensity. Finally, the maltose sensor with optimized linkers was redesigned to create a sugar sensor with a new specificity and wide dynamic range. When the optimized maltose sensors were employed as an in vivo sensor, highly-responsive FRET images were generated with application to the real-time analysis of maltose uptake in baker's yeast.