A functional proteomic analysis of the intracytoplasmic membrane (ICM) development process was performed in Rhodobacter sphaeroides during adaptation from high-intensity illumination to indirect diffuse light. This initiated an accelerated synthesis of the peripheral light-harvesting 2 (LH2) complex relative to that of LH1-reaction center (RC) core particles. After 11 days, ICM vesicles (chromatophores) and membrane invagination sites were isolated by rate-zone sedimentation and subjected to clear native gel electrophoresis. Proteomic analysis of gel bands containing the RC-LH1 and -LH2 complexes from digitonin-solubilized chromatophores revealed high levels of comigrating electron transfer enzymes, transport proteins, and membrane assembly factors relative to their equivalent gel bands from cells undergoing adaptation to direct low-level illumination. The GroEL chaperonin accounted for >65% of the spectral counts in the RC-LH1 band from membrane invagination sites, which together with the appearance of a universal stress protein suggested that the viability of these cells was challenged by light limitation. Functional aspects of the photosynthetic unit assembly process were monitored by near-IR fast repetition rate analysis of variable fluorescence arising from LH-bacteriochlorophyll a components. The quantum yield of the primary charge separation during the early stages of adaptation showed a gradual increase (variable/maximal fluorescence = 0.78-0.83 between 0 and 4 h), while the initial value of ~70 for the functional absorption cross section (σ) gradually increased to 130 over 4 days. These dramatic σ increases showed a direct relation to gradual slowing of the RC electron transport turnover rate (τ(QA)) from ~1.6 to 6.4 ms and an ~3-fold slowing of the rate of reoxidation of the ubiquinone pool. These slowed rates are not due to changes in UQ pool size, suggesting that the relation between increasing σ and τ(QA) reflects the imposition of constraints upon free diffusion of ubiquinone redox species between the RC and cytochrome bc(1) complex as the membrane bilayer becomes densely packed with LH2 rings.

Following platelet activation, platelets undergo a dramatic shape change mediated by the actin cytoskeleton and accompanied by secretion of granule contents. While the actin cytoskeleton is thought to influence platelet granule secretion, the mechanism for this putative regulation is not known. We found that disruption of the actin cytoskeleton by latrunculin A inhibited α-granule secretion induced by several different platelet agonists without significantly affecting activationinduced platelet aggregation. In a cell-free secretory system, platelet cytosol was required for α-granule secretion. Inhibition of actin polymerization prevented α-granule secretion in this system and purified platelet actin could substitute for platelet cytosol to support α-granule secretion. To determine whether SNAREs physically associate with the actin cytoskeleton, we isolated the Triton X-100 insoluble actin cytoskeleton from platelets. VAMP-8 and syntaxin-2 associated only with actin cytoskeletons of activated platelets. Syntaxin-4 and SNAP-23 associated with cytoskeletons isolated from either resting or activated platelets. When syntaxin-4 and SNAP-23 were tested for actin binding in a purified protein system, only syntaxin-4 associated directly with polymerized platelet actin. These data show that the platelet cytoskeleton interacts with select SNAREs and that actin polymerization facilitates α-granule release.The role of the actin cytoskeleton in granule exocytosis is enigmatic. It has been demonstrated to act both as a physical barrier that limits granule secretion and as a positive regulator of membrane fusion and cargo release. The ability of the resting actin cytoskeleton to serve as a barrier to granule secretion has been demonstrated in neutrophils, neurons, chromaffin cells, melanotrophs, pancreatic beta cells, and acinar cells (1)(2)(3)(4)(5)(6). We have previously demonstrated that platelet granules are coated with actin and that the actin cytoskeleton impedes platelet dense granule and α-granule release (7). Partial disruption of this barrier results in augmented and more rapid release of granule contents from platelets. This actin cytoskeletal barrier may help prevent unregulated release of thrombogenic substances into the circulation (7).Yet accumulating evidence indicates that actin polymerization can promote membrane fusion. Actin polymerization contributes to homotypic fusion of yeast vacuoles (8), fusion of phagosomes with endocytotic organelles (9) as well as secretion of granules from neuroendocrine cells (6,10,11), and mast cells (12). In some cells, actomyosin contraction and/ † Supported by NIH grants HL63250 and HL87203 (R.F.) and T32 HL07917 (K.W., L.D., P.S.B., C.G.P. (23,25,(27)(28)(29)(30).The dramatic morphologic change and granule centralization that occurs upon exposure to a strong agonist has lead to speculation that the cytoskeleton provides a contractile force that facilitates granule release (31)(32)(33). Yet this hypothesis remains unproven. An additional or alternative possibility is ...

Although the primary photochemical events in the facultative photoheterotrophic purple bacterium Rhodobacter sphaeroides are now well understood, comparatively little is known about how their photosynthetic apparatus is assembled. Here we present a proteomic analysis of the intracytoplasmic membrane (ICM) assembly process during adaptation to lowered light intensity, in which the size of the photosynthetic units is greatly expanded by addition of the light-harvesting 2 (LH2) peripheral antenna complex. When the isolated ICM-derived chromatophore vesicles were subjected to clear native gel electrophoresis (CNE), four pigmented bands appeared; the top and bottom bands contained the reaction center - light-harvesting 1 (RC-LH1) core complex and LH2 peripheral antenna, respectively, while the two bands of intermediate migration contained associations of the LH2 and core complexes. Proteomic analysis revealed a large array of other proteins associated with the CNE gel bands - in particular, several F(1)F(O)-ATP synthase subunits gave unexpectedly high spectral counts, given the inability to detect this coupling factor, as well as the more abundant cytochrome bc (1) complex, by atomic force microscopy. Significant levels of general membrane assembly factors were also found, as well as numerous proteins of unknown function including high counts for RSP6124 that were correlated with LH2 levels. When combined with further AFM and spectroscopic studies, these proteomic approaches are expected to provide a much-improved understanding of the overall assembly process.

Atomic force microscopy (AFM) of the native architecture of the intracytoplasmic membrane (ICM) of a variety of species of purple photosynthetic bacteria, obtained at submolecular resolution, shows a tightly packed arrangement of light harvesting (LH) and reaction center (RC) complexes. Since there are no unattributed structures or gaps with space sufficient for the cytochrome bc(1) or ATPase complexes, they are localized in membrane domains distinct from the flat regions imaged by AFM. This has generated a renewed interest in possible long-range pathways for lateral diffusion of UQ redox species that functionally link the RC and the bc(1) complexes. Recent proposals to account for UQ flow in the membrane bilayer are reviewed, along with new experimental evidence provided from an analysis of intrinsic near-IR fluorescence emission that has served to test these hypotheses. The results suggest that different mechanism of UQ flow exist between species such as Rhodobacter sphaeroides, with a highly organized arrangement of LH and RC complexes and fast RC electron transfer turnover, and Phaeospirillum molischianum with a more random organization and slower RC turnover. It is concluded that packing density of the peripheral LH2 antenna in the Rba. sphaeroides ICM imposes constraints that significantly slow the diffusion of UQ redox species between the RC and cytochrome bc(1) complex, while in Phs. molischianum, the crowding of the ICM with LH3 has little effect upon UQ diffusion. This supports the proposal that in this type of ICM, a network of RC-LH1 core complexes observed in AFM provides a pathway for long-range quinone diffusion that is unaffected by differences in LH complex composition or organization.