Photosynthesis converts absorbed solar energy to a protonmotive force, which drives ATP synthesis. The membrane network of chlorophyll–protein complexes responsible for light absorption, photochemistry and quinol (QH2) production has been mapped in the purple phototrophic bacterium Rhodobacter (Rba.) sphaeroides using atomic force microscopy (AFM), but the membrane location of the cytochrome bc1 (cytbc1) complexes that oxidise QH2 to quinone (Q) to generate a protonmotive force is unknown. We labelled cytbc1 complexes with gold nanobeads, each attached by a Histidine10 (His10)-tag to the C-terminus of cytc1. Electron microscopy (EM) of negatively stained chromatophore vesicles showed that the majority of the cytbc1 complexes occur as dimers in the membrane. The cytbc1 complexes appeared to be adjacent to reaction centre light-harvesting 1-PufX (RC-LH1-PufX) complexes, consistent with AFM topographs of a gold-labelled membrane. His-tagged cytbc1 complexes were retrieved from chromatophores partially solubilised by detergent; RC-LH1-PufX complexes tended to co-purify with cytbc1, whereas LH2 complexes became detached, consistent with clusters of cytbc1 complexes close to RC-LH1-PufX arrays, but not with a fixed, stoichiometric cytbc1-RC-LH1-PufX supercomplex. This information was combined with a quantitative mass spectrometry (MS) analysis of the RC, cytbc1, ATP synthase, cytaa3 and cytcbb3 membrane protein complexes, to construct an atomic-level model of a chromatophore vesicle comprising 67 LH2 complexes, 11 LH1-RC-PufX dimers & 2 RC-LH1-PufX monomers, 4 cytbc1 dimers and 2 ATP synthases. Simulation of the interconnected energy, electron and proton transfer processes showed a half-maximal ATP turnover rate for a light intensity equivalent to only 1% of bright sunlight. Thus, the photosystem architecture of the chromatophore is optimised for growth at low light intensities.

Reaction center-light harvesting 1 (RC-LH1) complexes are the fundamental units of bacterial photosynthesis, which use solar energy to power the reduction of quinone to quinol prior to the formation of the proton gradient that drives ATP synthesis. The dimeric RC-LH1-PufX complex of Rhodobacter sphaeroides is composed of 64 polypeptides and 128 cofactors, including 56 LH1 bacteriochlorophyll a (BChl a) molecules that surround and donate energy to the two RCs. The 3D structure was determined to 8 Å by X-ray crystallography, and a model was built with constraints provided by electron microscopy (EM), nuclear magnetic resonance (NMR), mass spectrometry (MS), and site-directed mutagenesis. Each half of the dimer complex consists of a RC surrounded by an array of 14 LH1 αβ subunits, with two BChls sandwiched between each αβ pair of transmembrane helices. The N- and C-terminal extrinsic domains of PufX promote dimerization by interacting with the corresponding domains of an LH1 β polypeptide from the other half of the RC-LH1-PufX complex. Close contacts between PufX, an LH1 αβ subunit, and the cytoplasmic domain of the RC-H subunit prevent the LH1 complex from encircling the RC and create a channel connecting the RC QB site to an opening in the LH1 ring, allowing Q/QH₂ exchange with the external quinone pool. We also identified a channel that connects the two halves of the dimer, potentially forming a long-range pathway for quinone migration along rows of RC-LH1-PufX complexes in the membrane. The structure of the RC-LH1-PufX complex explains the crucial role played by PufX in dimer formation, and it shows how quinone traffic traverses the LH1 complex as it shuttles between the RC and the cytochrome bc₁ complex.

No abstract

Macromolecular membrane assemblies of chlorophyll-protein complexes efficiently harvest and trap light energy for photosynthesis. To investigate the delivery of chlorophylls to the newly synthesized photosystem apoproteins, a terminal enzyme of chlorophyll biosynthesis, chlorophyll synthase (ChlG), was tagged in the cyanobacterium Synechocystis PCC 6803 (Synechocystis) and used as bait in pull-down experiments. We retrieved an enzymatically active complex comprising ChlG and the high-light-inducible protein HliD, which associates with the Ycf39 protein, a putative assembly factor for photosystem II, and with the YidC/Alb3 insertase. 2D electrophoresis and immunoblotting also provided evidence for the presence of SecY and ribosome subunits. The isolated complex contained chlorophyll, chlorophyllide, and carotenoid pigments. Deletion of hliD elevated the level of the ChlG substrate, chlorophyllide, more than 6-fold; HliD is apparently required for assembly of FLAGChlG into larger complexes with other proteins such as Ycf39. These data reveal a link between chlorophyll biosynthesis and the Sec/YidC-dependent cotranslational insertion of nascent photosystem polypeptides into membranes. We expect that this close physical linkage coordinates the arrival of pigments and nascent apoproteins to produce photosynthetic pigmentprotein complexes with minimal risk of accumulating phototoxic unbound chlorophylls.

Carotenoids protect the photosynthetic apparatus against harmful radicals arising from the presence of both light and oxygen. They also act as accessory pigments for harvesting solar energy, and are required for stable assembly of many light-harvesting complexes. In the phototrophic bacterium Rhodobacter (Rba.) sphaeroides phytoene desaturase (CrtI) catalyses three sequential desaturations of the colourless carotenoid phytoene, extending the number of conjugated carbon–carbon double bonds, N, from three to nine and producing the yellow carotenoid neurosporene; subsequent modifications produce the yellow/red carotenoids spheroidene/spheroidenone (N = 10/11). Genomic crtI replacements were used to swap the native three-step Rba. sphaeroides CrtI for the four-step Pantoea agglomerans enzyme, which re-routed carotenoid biosynthesis and culminated in the production of 2,2′-diketo-spirilloxanthin under semi-aerobic conditions. The new carotenoid pathway was elucidated using a combination of HPLC and mass spectrometry. Premature termination of this new pathway by inactivating crtC or crtD produced strains with lycopene or rhodopin as major carotenoids. All of the spirilloxanthin series carotenoids are accepted by the assembly pathways for LH2 and RC–LH1–PufX complexes. The efficiency of carotenoid-to-bacteriochlorophyll energy transfer for 2,2′-diketo-spirilloxanthin (15 conjugated CC bonds; N = 15) in LH2 complexes is low, at 35%. High energy transfer efficiencies were obtained for neurosporene (N = 9; 94%), spheroidene (N = 10; 96%) and spheroidenone (N = 11; 95%), whereas intermediate values were measured for lycopene (N = 11; 64%), rhodopin (N = 11; 62%) and spirilloxanthin (N = 13; 39%). The variety and stability of these novel Rba. sphaeroides antenna complexes make them useful experimental models for investigating the energy transfer dynamics of carotenoids in bacterial photosynthesis.