component t Fl ¼
3.3-3.4 ns stems from the I-state [ 78 , 79 ]. Subsequent compara-
tive investigations confirm that the missing hydrogen-bond prolongs t Fl in FPs [ 21 ].
The reason for this, at first glance, unexpected behaviour is of purely electronic
nature. The molecular mechanism, which is presumably responsible for the lifetime
shortening, is the one-bond rotation around the angle f. The anionic chromophore
can be described by two mesomeric forms, i.e., the quinoidal and the benzoidal
form. The double-bond character of the relevant exocyclic C-C bond is higher in
the quinoidal form, thereby stiffening the chromophore. In contrast, the benzoidal
form is strengthened when the negative charge is stabilized on the phenolate, e.g. by
hydrogen-bonding. Here, the chromophore can accommodate easier to the torque
pushing it away from planarity [ 38 ]. The same effect is observed when fluorinated
tyrosines are incorporated into the chromophore [ 76 ]: the negative charge is
stabilized because of the electron-withdrawing capability of fluorine atoms ( I
effect), which then leads to a reduction of t Fl despite strong interactions with the
surrounding in the protein (see Fig. 9b ).
3.3.2 Structural Effects of the Protein Matrix
The above described electronic effects certainly are most concrete and point to
the torsional deformation around f as major decay mechanism (see also Sect. 1.4 ).
This movement is made responsible for the fast excited-state decay of the synthetic
GFP chromophore and of BFPs [ 1 , 3 , 67 ]. Also, sophisticated quantum-mechanical
calculations confirm that an excited-state minimum is reached along this single-
bond rotation of the isolated anionic chromophore, thus reducing the energy gap
between S 0 and S 1 [ 37 ]. The effect of the protein matrix is to impose a strong
counterforce to this rotation [ 80 ]. A temperature dependence, which apparently
hints at an activated process, can be explained by the strong friction exerted by the
protein [ 81 ]. Therefore, lifetime measurements allow for conclusions about protein
dynamics [ 51 ].
There are also experimental data, which supports the importance of bulky aro-
matic amino acids such as phenylalanine or tyrosine at 203 for long t Fl . It seems that a
t Fl >
3.5 ns cannot be achieved in GFPs, i.e., proteins without aromatic residues at
203. In contrast, several authors reported t Fl close or even > 4 ns for YFPs, thereby
exceeding all published values for GFPs [ 21 , 82 , 83 ]. These values are close to the
theoretical limit t rad , which probably is ~4.6 ns [ 22 ] (see Fig. 2 ). One could argue
that the phenyl- or tyrosyl-moiety maintains the planarity of the chromophore.
Another important decay mechanism is the concerted two-bond flip around t
and f, i.e., the so-called Hula-twist (HT) [ 84 ]. It comprises at least four different
collective movements of atoms, with graduated importance [ 85 ]. That HT can
actually occur in FPs, can be inferred from some switching reactions, especially
of red FPs [ 86 , 87 ]. It is the only way how cis-trans isomerization can proceed
therein [ 38 ]. There is, however, no need of a completed isomerization reaction for
efficient fluorescence quenching [ 88 ].