We have recently completed studies which show that the saturation behavior for electrogenic proton pumping much more closely resembles that of the M-slow cycle than that of the M-fast cycle. Furthermore, the most electrogenic step in the overall photocycle has been identified as the decay of Ms to the ground state. These findings are completely compatible with the view that the photocycle is heterogeneous, based on different ground states. Our earlier studies have shown that the homogeneous BR in different membrane lipid environments displays different photocycle behaviors.? ? An alternative view of the photocycle, which is favored by many researchers, is that of a single, homogeneous cycle with reversible steps as in? ? BR<-->L1<-->L2<-->M1<-->M2<-->N<-->O-->BR.? ? This scheme, as it stands, is incompatible with a large number of experimental observations indicating heterogeneity, such as the different decay paths for the two forms of M, their differences in actinic light saturation, and their different proton-pumping abilities as they return to the ground state. Adherents of the single, homogeneous, reversible cycle model maintain that all such apparent inconsistencies can be reconciled by photocooperativity. BR, in the membrane, exists in trimer units. It is proposed that photocycles initiated by multiple photon hits to the target are different than those ensuing after a single hit. We have experimentally tested several of the most prominent proposed kinetic models based on photocooperativity by comparing their predicted photocycles with those we have measured under a very wide range of increasing actinic energies of laser flashes. We find that all proposed models fail, most notably because: 1) predictions are based on 2 and 3 monomer turnovers per trimer whereas the amount of turnover reaches a plateau at a level where only 1 of the 3 monomers per trimer turns over; 2) the models predict that at very high energy levels, all Mf should decrease to 0 as Ms rises to a maximum, whereas both actually reach steady saturation plateaus; 3 the different decay paths and saturation behaviors that are observed are not accounted for. On the other hand, a mathematical model based on heterogeneous ground states with no photocooperativity closely resembles the experimental data. ? ? Our approach to defining the conformational changes in BR that occur during electrogenic proton movements is based on the linear algebra-based deconvolution procedures we introduced which allows one to derive absolute optical spectra for each intermediate in the photocycle. We are starting with the application of FTIR. At NIST, we have two optical benches, side-by-side, with our time-resolved optical spectrometer on one and an IR spectrometr on the other. We have designed, built, and tested an arrangement which performs visible spectroscopy on the same CaFl2 disc windows used in the FTIR studies. We plan to combine the optical and FTIR data in one matrix and apply our deconvolution procedures to isolate both optical and IR absolute spectra for each intermediate. With this approach, we expect to find new information on conformational and proton-binding changes occurring at the most electrogenic steps in the photocycle.? A new collaborative initiative with the National Institute for Biomedical Imaging and Bioengineering was started. The object is to build a spectrometer capable of performing rapid optical and FTIR kinetic measurements on single membrane-protein crystals as well as tiny fragments of intact membrane, containing the in-situ protein. This instrument will serve several vital functions: 1.) quality control to determine how closely the isolated crystal reproduces the function of the protein in the membrane. 2.) to serve as a guide whereby the preparation of the crystal can be modified to form more native-like functions. 3.) to possibly lead to isolation of 3D molecular structures for each of the intermediates in a kinetic sequence such as electrogenic proton pumping by BR. The latter goal is envisaged by using validated membrane protein crystals in time-resolved X-ray crystallography, and our mathematical procedures which lead to isolation of pure visible spectra to isolate unique diffraction patterns for the intermediates and from these to obtain the atomic x, y, and z coordinate values. If successful, this should provide precise information on the molecular rearrangements of a protein as it transduces energy from a proton current into an electrochemical gradient.
