Photophysical Engineering of Fluorescent Proteins: Accomplishments and Challenges of Physical Chemistry Strategies

Abstract

Fluorescent proteins (FPs) have become ubiquitous tools for biological research and concomitantly they are intriguing molecules that are amenable to study with a wide range of experimental and theoretical tools. This perspective explores the connection between the engineering of improved FPs and basic ideas from physical chemistry that explain their properties and drive the molecular design of brighter and more photostable variants. We highlight some of the progress and the many knowledge gaps in understanding the relationship between FP brightness and photostability. We also explore some of the pertinent remaining questions and suggest ways in which physical chemists might further examine the physical basis of brightness and photostability in these systems.

Directed evolution of a bright variant of mCherry: Suppression of non-radiative decay by fluorescence lifetime selections

“…The fluorescence quantum yields of these FPs show a clear correlation with the average fluorescence lifetime (SI Figure S12a), though the trend is not perfectly linear. 5,8 The two-fold decrease in the half-life of photobleaching with a 2.5-fold increase in fluorescence lifetime also follow expected trends, agreeing with observations of higher photobleaching for FPs with longer fluorescence lifetimes (SI Figure S6 & S7). 5, 8 Additionally, a linear fit (Radj 2 =0.95) of the lifetime and quantum yield resulted in a slope of 150 ± 14 µs -1 .…”

“…The value of kr can be explained by the Strickler-Berg equation, which relates radiative rate to the peak extinction coefficient and peak fluorescence frequency. 8,30 The small variation in kr is further corroborated by the observation that the peak extinction coefficient for this series only varies 15% from an average value of εmax Avg ~70000 M -1 cm -1 . This variation is modest…”

“…This frequency dependence of the radiative rate, predicted by the Strickler-Berg equation, is expected to yield a linear relationship for chromophores with relatively narrow fluorescence spectra (Supplementary Figure S12c). 8,30,31 Most significantly, we found the value of knr undergoes the largest change (a 6.5-fold decrease) across this series. A consistent blue-shift in the absorption and emission peak wavelengths with higher brightness, correlated with this decrease in knr led us to investigate models that could explain this observation.…”

“…Although it is widely acknowledged that lifetime and other photophysical properties such as photobleaching or spectral changes frequently co-evolve, a mechanistic investigation of their interdependence has rarely been pursued. 8 In the GFP superfamily, the chromophore is formed when a tripeptide of the internal α-helix embedded in an 11-stranded β-barrel undergoes rearrangement, cyclization, dehydration, and oxidation. 9,10 This process results in a π-conjugated structure of p-hydroxybenzylidine and imidazolinone rings bridged through a methine carbon.…”

Photoswitchable Fluorescent Proteins: Mechanisms on Ultrafast Timescales

“…Studies on photoisomerization events have been a hot topic for the HBDI chromophore and its derivatives in solution, but the gained knowledge cannot be directly transferred into a chromophore in the protein matrix due to the complex electrostatic and steric interactions between the chromophore and surrounding protein residues, which are intrinsically more heterogeneous and asymmetric than the solution environment [ 39 , 40 , 41 , 42 ]. Therefore, it is urgent and necessary to elucidate the detailed structural motions of the chromophore and local amino acids in the electronically excited and “hot” (non-equilibrium) ground states, which are essential for the deepened understanding and rational development of next-generation RSFPs [ 43 ].…”