Abstract
Resonance Raman spectra of bacteriorhodopsin (in various isotopically labelled and natural environments) and rhodopsins demonstrate that in all bacteriorhodopsin species with protonated Schiff base linkages, there is a secondary protein-Schiff base interaction. Steady state resonance Raman spectra of 15 N labelled purple membrane fragments and kinetic resonance Raman spectra (k.r.R.s.) as a function of pH, are consistent with the suggestion that the e-amino group of a lysine residue is responsible for this secondary protein-protonated Schiff base interaction. Spectra of bacteriorhodopsin photo-intermediates with unprotonated Schiff base linkages demonstrate that the secondary protein-Schiff base interaction is absent in these species. These results, together with the accessibility of the protonated Schiff base in bacteriorhodopsin (bR 570 ) to solvent from the external medium and the insensitivity to pH of the secondary lysine-Schiff base complex in bR 570 , suggest a structure for the complex in which the Schiff base proton is interacting with an amino acid side chain in an unprotonated configuration. In addition k.r.R.s. as a function of pH has demonstrated that the protein-Schiff base complex has a p K > 12 before light absorption and a p K in the range between 9.9 and 10.3 microseconds after light absorpion. These results suggest a molecular mechanism for proton pumping, which appears to account for the changes that are observed in the visible absorption spectrum and the resonance Raman spectrum as a function of proton pumping and pH. Finally, our results on bacteriorhodopsin and rhodopsin indicate that the primary excitation mechanism in these pigments, which produce remarkably similar absorption red shifts on similar time scales with different chromophore conformations and Schiff base interactions, must involve some region in the active site removed from the Schiff base and unrelated to a simple 11- cis to all- trans isomerization. A proton translocation in the protein together with retinal structural alteration (Lewis 1978) certainly fits all our observations and results in plausible molecular transformations which can account for the spectral similarities of all rhodopsins and bacteriorhodopsin while accounting for their functional diversity.