Structural and functional analysis of the reaction center-light harvesting complex of Rhodobacter sphaeroides

Dimeric photosynthetic reaction center-light harvesting complex (RC-LH1) has been purified from the photosynthetic bacterium Rhodobacter sphaeroides grown under semiaerobic conditions. Removal of the detergent in the presence of lipids leads to the formation of two-dimensional crystals. Analysis by cryoelectron mycroscopy at a resolution of 26 A ° reveals an ‘‘S’’-shaped dimeric complex where the continuity of the LH1 ring that surrounds the RC is interrupted. The higher density of the projection map at the junction between the two monomers of core complex is attributed to a dimer of the PufX peptide (Scheuring et al., 2004, J. Biol. Chem. 279, 3620). These data confirm the structural role of PufX, a single transmembrane protein required for the photosynthetic phenotype (Farchaus et al., 1992, EMBO J., 11, 2779), responsible for the dimerization of the RC-LH1 complex (Francia et al., 1999, Biochemistry 38, 6834).The functionality of the isolated complex, purified from photosynthetically and semiaerobically grown bacteria, was analyzed by time resolved spectroscopy. Upon excitation of the sample with an actinic laser pulse, the kinetics of charge recombination from the state P +QAQB- to the neutral state PQAQB exibhit a slow phase with an half time of approximately 4 s, at least four times larger than what usually observed in RC complex deprived of the LH1. Stoichiometric determinations of the quinone (Q10) present in the RCLH1 indicate a Q10/RC-LH1 ratio >10. These quinones are functionally coupled to the RC-LH1 complex, as judged from the extent of cytochrome c2 rapidly oxidized under continuous illumination. Charge recombination kinetics have been analyzed on the basis of a model proposed by Shinkarev and Wraight (Shinkarev and Wraight, 1993, in ‘‘The photosynthetic reaction center’’ Vol. 1, 193) that take into account the binding of quinone at the QB site when a quinone pool is present. The slowing down of the recombination reaction experimentally detected cannot be simply explained by a quinone concentration effect. The model predicts a lower limit of the charge recombination rate constant when the quinone concentration is raised to infinity that is well above the one measured in the RC-LH1, indicating that, even in the presence of saturating quinone conditions the recombination reaction cannot be so slow as experimentally observed. These data suggests that a stabilization of the charge separated state P +QAQB-, leading to a slower recombination reaction, is induced by the LH1 antenna complex.