Comparing the Performance of mScarlet-I, mRuby3, and mCherry as FRET Acceptors for mNeonGreen

Abstract Förster Resonance Energy Transfer (FRET) has become an immensely powerful tool to profile intra- and inter-molecular interactions. Through fusion of genetically encoded fluorescent proteins (FPs) researchers have been able to detect protein oligomerization, receptor activation, and protein translocation among other biophysical phenomena. Recently, two bright monomeric red fluorescent proteins, mRuby3 and mScarlet-I, have been developed. These proteins offer much improved physical properties compared to previous generations of monomeric red FPs that should help facilitate more general adoption of Green/Red FRET. Here we assess the ability of these two proteins, along with mCherry, to act as a FRET acceptor for the bright, monomeric, green-yellow FP mNeonGreen using intensiometric FRET and 2-photon Fluorescent Lifetime Imaging Microscopy (FLIM) FRET techniques. We first determined that mNeonGreen was a stable donor for 2-photon FLIM experiments under a variety of imaging conditions. We then tested the red FP’s ability to act as FRET acceptors using mNeonGreen-Red FP tandem construct. With these constructs we found that mScarlet-I and mCherry are able to efficiently FRET with mNeonGreen in spectroscopic and FLIM FRET. In contrast, mNeonGreen and mRuby3 FRET with a much lower efficiency than predicted in these same assays. We explore possible explanations for this poor performance but are unable to definitively determine the cause, all though protein maturation seems to play a role. Overall, we find that mNeonGreen is an excellent FRET donor, and both mCherry and mScarlet-I, but not mRuby3, act as practical FRET acceptors, with mScarlet-I out performing mCherry due it’s higher brightness.