Far infrared (FIR) spectral measurements of wild-type (WT) and D96N mutant bacteriorhodopsin thin films have been carried out using terahertz time domain spectroscopy as a function of hydration, temperature, and conformational state. The results are compared to calculated spectra generated via normal mode analyses using CHARMM. We find that the FIR absorbance is slowly increasing with frequency and without strong narrow features over the range of 2-60 cm(-1) and up to a resolution of 0.17 cm(-1). The broad absorption shifts in frequency with decreasing temperature as expected with a strongly anharmonic potential and in agreement with neutron inelastic scattering results. Decreasing hydration shifts the absorption to higher frequencies, possibly resulting from decreased coupling mediated by the interior water molecules. Ground-state FIR absorbances have nearly identical frequency dependence, with the mutant having less optical density than the WT. In the M state, the FIR absorbance of the WT increases whereas there is no change for D96N. These results represent the first measurement of FIR absorbance change as a function of conformational state.

The structural changes that accompany activation of a G-protein coupled receptor (GPCR) are not well understood. To better understand the activation of rhodopsin, the GPCR responsible for visual transduction, we report studies on the three-dimensional structure for the activated state of this receptor, metarhodopsin II. Differences between the three-dimensional structure of ground state rhodopsin and metarhodopsin II, particularly in the cytoplasmic face of the receptor, suggest how the receptor is activated to couple with transducin. In particular, activation opens a groove on the surface of the receptor that could bind the N-terminal helix of the G protein, transducin alpha.

The absorption maximum of blue proteorhodopsin (BPR) is the most blue-shifted of all retinal proteins found in archaea or bacteria, with the exception of sensory rhodopsin II (SRII). The absorption spectrum also exhibits a pH dependence larger than any other retinal protein. We examine the structural origins of these two properties of BPR by using optical spectroscopy, homology modeling, and molecular orbital theory. Bacteriorhodopsin (BR) and SRII are used as homology parents for comparative purposes. We find that the tertiary structure of BPR based on SRII is more realistic with respect to free energy, dynamic stability, and spectroscopic properties. Molecular orbital calculations including full single- and double-configuration interaction within the chromophore pi-electron system provide perspectives on the wavelength regulation in this protein and indicate that Arg-95, Gln-106, Glu-143, and Asp-229 play important, and in some cases pH-dependent roles. A possible model for the 0.22 eV red shift of BPR at low pH is examined, in which Glu-143 becomes protonated and releases Arg-95 to rotate up into the binding site, altering the electrostatic environment of the chromophore. At high pH, BPR has spectroscopic properties similar to SRII, but at low pH, BPR has spectroscopic properties more similar to BR. Nevertheless, SRII is a significantly better homology model for BPR and opens up the question of whether this protein serves as a proton pump, as commonly believed, or is a light sensor with structure-function properties more comparable to those of SRII. The function of BPR in the native organism is discussed with reference to the results of the homology model.

The absorption spectrum of green proteorhodopsin (GPR) is pH-dependent, exhibiting either red-shifted (low pH) or blue-shifted (high pH) absorption maxima. We examine the molecular basis of the pH-dependent spectral properties of green proteorhodopsin by using homology modeling and molecular orbital theory. Bacteriorhodopsin (BR) and sensory rhodopsin II (SRII) are compared as homology templates. The model of GPR generated by using BR as the homology parent is better than that generated by using SRII on the basis of the potential energy, relative stability to dynamics, and ability to rationalize pH effects. MNDO-PSDCI (molecular neglect of differential overlap with partial single- and double-configuration interaction) calculations provide insight into the spectroscopic properties of GPR and help rule out the viability of the SRII-based model. The proximity of His 75 to the quadrupole residues (LYR, D97, D227, and R94) in the BR-based model provides a good model for both the low- and high-pH spectral states of GPR. The observation that BR is a better structural model for GPR than SRII is in contrast to our previous study of BPR, which observed that SRII was the better homology parent [Hillebrecht, J. R. (2006) Biochemistry 45, 1579-1590]. The implications of this observation are discussed.