Fluorescent protein resonance energy transfer for high-throughput screening

The desire to track proteins and monitor their functions inside of living cells has driven the development of new tools and techniques capable of making these measurements. Fluorescent protein Forster resonance energy transfer (FRET) is an attractive choice for probing intracellular events, but the low efficiency of most available pairs limits the utility of the technique. To address this problem, a quantitative evolutionary strategy using fluorescence-activated cell sorting was applied to optimize a cyan-yellow fluorescent protein pair for FRET. Multiple rounds of mutation, followed by sorting of the leading edge of the cell population enabled step-wise evolution of variants with improved FRET properties. The best pair of variants, CyPet-YPet, exhibited a seven-fold improvement in dynamic range over the parental pair when measured in vitro. Additionally, the optimized pair provided significantly improved resolution of FRET signal changes in bacterial, yeast, and mammalian cells using flow cytometry. The improvements in dynamic range were verified in applications including the detection of protease activities, protein-protein interactions, and small molecules.; The improved utility of CyPet and YPet enabled flow cytometric sorting on the basis of fluorescent protein FRET for the first time. Clones expressing active caspase-3 were isolated from inactive background present in 10 5-fold excess using the FRET signal change as an indicator of protease function. This proof of principle experiment paves the way for new opportunities in the isolation and evolution of potentially therapeutic proteases. Additionally, a FRET-based protein-protein interaction screen was developed, and enabled enrichment of peptide binders to an SH3 domain and a PDZ2 domain from large libraries. Consensus sequences determined from isolated peptides matched the known consensus binding motifs for these targets. This FRET approach may have advantages over yeast two-hybrid and other screens. Additionally, flow cytometric sorting techniques developed for isolating rare cells on the basis of small FRET signal changes may have more widespread applicability to situations where target cell fluorescence differs subtly from that of nontarget cells. These tools and techniques developed for fluorescent protein FRET applications create opportunities for expansion of the utility of FRET in the discovery and study of biological interactions and activities.