Advantages of substituting bioluminescence for fluorescence in a resonance energy transfer-based periplasmic binding protein biosensor

A genetically encoded maltose biosensor was constructed, comprising maltose binding protein (MBP) flanked by a green fluorescent protein (GFP(2)) at the N-terminus and a Renilla luciferase variant (RLuc2) at the C-terminus. This Bioluminescence resonance energy transfer(2) (BRET2) system showed a 30% increase in the BEET ratio upon maltose binding, compared with a 10% increase with an equivalent fluorescence resonance energy transfer (FRET) biosensor. BRET2 provides a better matched Forster distance to the known separation of the N and C termini of MBP than FRET. The sensor responded to maltose and maltotriose and the response was completely abolished by introduction of a single point mutation in the BRET2 tagged MBP protein. The half maximal effective concentration (EC50) was 0.37 mu M for maltose and the response was linear over almost three log units ranging from 10 nM to 3.16 mu M maltose for the BRET2 system compared to an EC50 of 2.3 mu M and a linear response ranging from 0.3 mu M to 21.1 mu M for the equivalent FRET-based biosensor. The biosensor's estimate of maltose in beer matched that of a commercial enzyme-linked assay but was quicker and more precise, demonstrating its applicability to real-world samples. A similar BRET2-based transduction scheme approach would likely be applicable to other binding proteins that have a "venus-fly-trap" mechanism. (C) 2012 Elsevier B.V. All rights reserved.