Structure–Function Relationships in Fluorescent Marker Proteins of the Green Fluorescent Protein Family - Fluorescent Proteins: From Understanding to Design - page 248

p -electron system. This reaction creates the red 2-[(1E)-2-(5-imidazolyl)ethenyl]-

4-( p -hydroxybenzylidene)-5-imidazolinone chromophore [ 85 ]. Nienhaus et al.
proposed an ESPT from the hydroxyl group of Tyr63 to the N

e of His62 to generate

the doubly protonated His62, followed by the -elimination step in which Glu212

acts as a proton acceptor [ 81 ]. Hayashi et al. proposed a water-assisted mechanism
to explain the loss of a water molecule (W1) in the red form of Kaede [ 86 ], which is
also observed in EosFP [ 81 ]. They all agreed that His62 is an essential group in the
conversion mechanism, which is underscored by the observation that EosFP [ 81 ]
and also Kaede [ 85 ] lose their photoconversion abilities after replacement of His62
by other amino acids [ 52 ]. Moreover, there was a general agreement that the neutral
hydroxyphenyl is the reactive species, which loses its proton upon excitation.
Recently, a theoretical study has suggested a mechanism in which the hydroxyphe-
nyl moiety of the chromophore remains protonated, and there is an ESPT from
His62 to Phe61 that promotes peptide bond cleavage [ 87 ]. It is impressive to see
how this intricate covalent chemistry takes place in the interior of the EosFP

b

b -can

without causing any further changes in the protein structure.
Photoconvertible proteins of this class have found widespread use as “pulse-
chase” labels in live-cell experiments [ 57 , 82 , 88 ]. Activated molecules, turned red
by photoactivation, yield bright red-fluorescent signals that are spectrally well
separated from the signals by FPs that have not been photoconverted. Especially,
these measurements are not adversely affected by freshly synthesized FP molecules
because those will emit in the green range of the spectrum. Targeted photoconver-
sion thus enables spatial and temporal marking of specific structures and tracking of
their signals in time in living cells.

Reversible Photoswitching Between Bright and Dark States

In 2000, Lukyanov et al. found a nonfluorescent homolog of GFP in the sea
anemone A. sulcata [ 28 ] that, upon illumination with green light, emitted increasing
intensities of red fluorescence at 595 nm. Therefore, it was termed asFP595. The
mutation Ala148Gly improved the “kindling” properties [ 89 ]; this variant was
called KFP-1 [ 89 ]. An entirely new generation of improved optical highlighters
with reversible on-off switching capabilities was introduced with the photoswitch-
able FP Dronpa [ 11 ]. By excitation with light of two different wavelengths, the
protein can be toggled between a bright fluorescent and a dark state. The structural
basis of this reversible photoswitching has been elucidated by crystallographic
studies of several FPs including Dronpa [ 90 , 91 ], mTFP0.7 [ 92 ], and asFP595
[ 29 , 31 , 93 ]. Under equilibrium conditions, the chromophores of Dronpa and
mTFP0.7 assume the cis conformation, with the anionic state of the chromophore
being the fluorescent species. Photoactivation of the anionic chromophore triggers a
cis - trans photoisomerization accompanied by a change of the protonation state. As
a result, the chromophore becomes essentially nonfluorescent. In KFP1 and
asFP595, the nonfluorescent trans state is thermodynamically more stable than
the cis state, and light irradiation induces a trans-cis isomerization to the fluores-
cent cis state. When kept in the dark, all FPs relax within minutes to several hours to