We considered a series of model systems for treating the photoabsorption of the 11-cis retinal chromophore in the protonated Schiff-base form in vacuum, solutions, and the protein environment. A high computational level, including the quantum mechanical-molecular mechanical (QM/MM) approach for solution and protein was utilized in simulations. The S0-S1 excitation energies in quantum subsystems were evaluated by means of an augmented version of the multiconfigurational quasidegenerate perturbation theory (aug-MCQDPT2) with the ground-state geometry parameters optimized in the density functional theory PBE0/cc-pVDZ approximation. The computed positions of absorption bands lambdamax, 599(g), 448(s), and 515(p) nm for the gas phase, solution, and protein, respectively, are in excellent agreement with the corresponding experimental data, 610(g), 445(s), and 500(p) nm. Such consistency provides a support for the formulated qualitative conclusions on the role of the chromophore geometry, environmental electrostatic field, and the counterion in different media. An essentially nonplanar geometry conformation of the chromophore group in the region of the C14-C15 bond was obtained for the protein, in particular, owing to the presence of the neighboring charged amino acid residue Glu181. Nonplanarity of the C14-C15 bond region along with the influence of the negatively charged counterions Glu181 and Glu113 are found to be important to reproduce the spectroscopic features of retinal chromophore inside the Rh cavity. Furthermore, the protein field is responsible for the largest bond-order decrease at the C11-C12 double bond upon excitation, which may be the reason for the 11-cis photoisomerization specificity.

Photoabsorption spectra of deprotonated trans p-coumaric acid and two of its methyl substituted derivatives have been studied in gas phase both experimentally and theoretically. We have focused on the spectroscopic effect of the location of the two possible deprotonation sites on the trans p-coumaric acid which originate to either a phenoxide or a carboxylate. Surprisingly, the three chromophores were found to have the same absorption maximum at 430 nm, in spite of having different deprotonation positions. However, the absorption of the chromophore in polar solution is substantially different for the distinct deprotonation locations. We also report on the time scales and pathways of relaxation after photoexcitation for the three photoactive yellow protein chromophore derivatives. As a result of these experiments, we could detect the phenoxide isomer within the deprotonated trans p-coumaric acid in gas phase; however, the occurrence of the carboxylate is uncertain. Several computational methods were used simultaneously to provide insights and assistance in the interpretation of our experimental results. The calculated excitation energies S(0)-S(1) are in good agreement with experiment for those systems having a negative charge on a phenoxide moiety. Although our augmented multiconfigurational quasidegenerate perturbation theory calculations agree with experiment in the description of the absorption spectrum of anions with a carboxylate functional group, there are some puzzling disagreements between experiment and some calculational methods in the description of these systems.

Green fluorescent protein, GFP, has revolutionized biology, due to its use in bioimaging. It is widely accepted that the protein environment makes its chromophore fluoresce, whereas the fluorescence is completely lost when the native chromophore is taken out of GFP. By the use of a new femtosecond pump-probe scheme, based on time-resolved action spectroscopy, we demonstrate that the isolated deprotonated GFP chromophore can be trapped in the first excited state when cooled to 100 K. The trapping is shown to last for 1.2 ns, which is long enough to establish conditions for fluorescence and consistent with calculated trapping barriers in the electronically excited state. Thus, GFP fluorescence is traced back to an intrinsic chromophore property, and by improving excited-state trapping, protein interactions enhance the molecular fluorescence.