The shift of the absorption maximum of a protonated retinal Schiff base upon binding to the apoprotein, the so-called opsin-shift, is the most conspicuous evidence of the special interaction between the chromophore and its environment in the protein binding pocket. Tuning the sensitivity of the visual pigment to different spectral regions is necessary for survival in different habitats; it is also prerequisite for color vision. Mechanisms which have been proposed in the past to explain the origin of the opsin shift could not be tested properly, since the structures were largely hypothetical and computational tools for quantitative calculation of the spectra were not available. With the recent X-ray structures of retinal binding proteins and the development of reliable quantum mechanical methods the calculation of excited states of the retinal chromophore including parts of the protein binding pocket has become feasible. We propose the non-empirical calculation of the excited states of chromophore models of retinal binding proteins on the basis of CASPT2 corrected multiconfigurational SCF wavefunctions. Chromophore geometries will be optimised subject to the constraints of the known X-ray data. The environment will be included in the calculations. On the basis of these calculations we will analyse different mechanisms of the opsinshift and evaluate their relative contributions. The calculations will support, /via/ the UV/Vis spectra, the structure elucidation of the intermediates of the rhodopsin and the bR photocycles.

Rhodopsin. Three factors have been identified which contribute to the opsin shift in the dark state of the protein, namely the internal distortion of the chromophore caused by steric interaction within the protein pocket; the effect of the charged counterion; and the polar environment due to the amino acids close to the chromophore. Based on the quantum-mechanically refined model of the rhodopsin binding pocket and employing CASPT2-methodology quantitative agreement between calculated and experimental wavelength shift of the chromophore was obtained. Contrary to theoretical evidence presented elsewhere we find the largest contributor to be the counterion, with the protein environment mainly responsible for maintaining the photochemically "alert" geometry of the chromophore and fine-tuning the absorbance.

Bathorhodopsin. With the advent of the crystal structure for the first intermediate following photoexcitation of rhodopsin the technology developed for the analysis of the rhodopsin state has been applied and found to yield a very reasonable bathochromic shift with respect to the dark state. More x-ray information is coming out regarding later states, such as lumirhodopsin and metarhodopsin I, where the theoretical analysis may be of use to validate the proposed structural models.

Isorhodopsin. The structure of this important rhodopsin analogue which shares the same photointermediate with the wild type, has been solved recently. We will use the small blue-shifted absorbance of isorhodopsin (lmax 485 nm) compared to rhodopsin (498 nm) as an indicator of the chromophore conformation and its interaction with the protein environment in order to verify the X-ray structure or suggest changes.