ESPT, or even better light activated proteins. Moreover, these experiments set the
scene for using GFP mutants as a model system for understanding the molecular
basis for proton transfer in biological systems.
While the previously cited experiments revealed the final proton acceptor for the
ESPT in wt GFP, they could not determine what was the trigger for the reaction,
which was considered the rate-limiting step. The only speculation was that the
observed strong coupling of the electronic excitation to the protein structure may
act as the trigger for the proton-transfer reaction.
Some more information about this arose from a different technique, the FSRS. In
their published FSRS study [ 56 ], Mathies and coworkers could determine a mode at
280 fs (120 cm 1 ) in the excited state of wt GFP just before the ESPT. An example
of the collected data is shown in Fig. 8a , together with the assignment of the
chromophore modes that mostly contribute in the given spectral region. The A *
spectrum appears with a 120-fs rise time and decays to I * on the picoseconds
timescale. Superimposed to this decay, there is an oscillation of the frequencies
of some of the modes, as highlighted in Fig. 8b . In particular, both the frequencies
and intensities (data not shown, see [ 56 ]) of two marker bands, the C-O and C
stretching modes at opposite ends of the conjugated chromophore, oscillate out of
phase with a period of 280 fs, indicating an out-of-phase oscillating behavior in the
length and bond order of these two bonds (similarly to what discussed in Sect. 4.2 ).
All these observations can be explained by the impulsively excited low frequency
phenoxyl-ring wagging motions shown in Fig. 8c ; this assignment is confirmed by
the fact that in DFT calculations in the electronic ground state, a two-ring out-
of-plane wagging mode of the chromophore was identified with a similar frequency
of 110 cm 1 . This motion was identified as the trigger for the occurrence of
ESPT by optimizing the geometry of the chromophore for the latter.
5 Conclusions and Perspectives
Vibrational spectroscopy can provide detailed information on molecular structure,
orientation or interaction with neighboring species; in this way, it is similar to
structural nuclear magnetic resonance (NMR). While the connections between the
structure in a molecule and its vibrational spectrum are less straightforward than
in the case of NMR signatures, IR and Raman spectroscopy have the advantage
that they can be applied to much bigger specimens; moreover, carefully exploiting
the (pre-)resonance conditions, Raman spectroscopy can address the structure of
parts of a macromolecule.
In the previous sections, we have reviewed how structural changes in GFP and
its mutants have been analyzed using vibrational spectroscopy. This process
required first a careful assignment of the vibrational modes of the chromophore in
these fluorescent proteins to each peak in their IR or Raman spectra. This assign-
ment has been carried out mostly by comparison with the theoretical simulations
described in Sect. 3 or by studying the spectral similarities and differences with