PRIMARY PROCESSES AND STRUCTURE OF THE PHOTOSYSTEM-II REACTION-CENTER.3. KINETIC-ANALYSIS OF PICOSECOND ENERGY-TRANSFER AND CHARGE SEPARATION PROCESSES IN THE D1-D2-CYT-B559 COMPLEX MEASURED BY TIME-RESOLVED FLUORESCENCE

An extensive picosecond fluorescence study has been performed on isola ted D1-D2 reaction centers (RCs) from photosystem (PS) II at room temp erature as a function of excitation and emission wavelength with a tim e resolution of about 2 ps. The study comprizes more than 50 excitatio n/emission wavelength pairs. Global analysis for individual excitation wavelengths and, alternatively, for the entire excitation/emission wa velength matrix has been performed. These analyses show that over a ti me range of 2 ns five exponentials are required to adequately describe the data. The shortest lifetime component, whose amplitude dominates the excited state decay for excitation in the red part of the absorpti on spectrum, is approximate to 3 ps. Further lifetime components fall into the 6-12, 20-30, and 60-130 ps and the several nanosecond ranges. In the combined global analysis at least five lifetime components (40 0 ps fit window), again with approximate to 3 ps as the shortest one ( again dominant in amplitude for long excitation wavelengths), and most likely even six components are required to describe the data. The dom inance and necessity of the approximate to 3 ps component is demonstra ted also by exhaustive search error analysis on the data in the global mode. Kinetic target analysis on models of various complexity has bee n performed, again on data sets from individual excitation wavelengths and on the combined data set. Such analyses have been performed for t he first time for the D1-D2 kinetics, and they allow an assignment of the origin of the various lifetime components. The target analyses yie lded realistic species associated spectra (SAS), as expected for Chi e xcited states and also the rate constants for the energy transfer and charge separation processes. The results show that (i) the majority of the effective primary charge separation is associated with the approx imate to 3 ps lifetime, (ii) three pools of external Chls, transferrin g their energy relatively slowly to the RC core, are attached to the D 1-D2 RC, (iii) the three Chls have energy transfer times in the range of 6-30 ps, and (iv) a approximate to 50 ps component is associated wi th a transition between different radical pairs. We exclude on the bas is of these data that the primary charge separation could occur primar ily in 21 ps, as proposed by other authors (Durrant; et al. Biochemist ry 1993, 32, 8259-8267). In all models, quite independent of their com plexity, the effective primary charge separation rate (from the equili brated RC core) is in the range of 100-150 ns(-1). If one accounts for the distribution of the excited state in the RC core over various chr omophores an intrinsic charge separation rate constant from P680(doubl e dagger) to the primary radical pair can be estimated to be approxima te to 360 ns(-1) corresponding to an intrinsic charge separation lifet ime of approximate to 2.7 ps, in very good agreement with our previous extrapolation based on the kinetics of intact PS II core particles (S chatz; et al. Biophys. J. 1988, 54, 397-405). It is concluded that the intrinsic primary charge separation in isolated D1-D2 reaction center s has about the same rate constant as in intact photosystem II.