Fig. 1. Size distributions of chromatophores from R. capsulatus. The distribution of vesicle diameters in a chromatophores preparation from photosynthetically grown R. capsulatus was analyzed. Images were obtained by cryo (empty bins, 462 vesicles) and negative staining (filled bins, 481 vesicles) EM.

Fig. 2. Bchl concentration profile along the sucrose density gradient. Distribution of Bchl concentration in fractions collected after sucrose density gradient under three different centrifugation conditions. Separation was performed using either a fixed angle rotor (Beckman 50.2Ti) at 45,000 rpm for 4 h (full circles) or a vertical rotor (Beckmann VTi 65.1) at 50,000 rpm for either 1 h or 12 h (full and empty triangles, respectively). The Bchl maximum was reproducibly found in fractions between 32.5% and 33.5% sucrose (w/w). The combined average sucrose concentration calculated from the three distribution is 34.1, corresponding to an average density of 1.146 g cm^− 3.

Fig. 3. Size distribution of chromatophores in different density fractions. The distribution of chromatophore diameters in three different density fractions was analyzed by negative staining EM. For the 30%, 35% and 39% fractions, 393, 377 and 362 vesicles were measured, respectively.

Fig. 4. ATP synthase activities of chromatophores along the sucrose gradient. (A) The Bchl distribution in a sucrose density gradient is reported (full circles) along with the hydrolytic activity in the presence of 0.5% LDAO (full triangle) and the synthesis activity measured by acid–base transition (empty circles). (B) The specific hydrolysis and synthesis activities on a Bchl basis, plotted against sucrose concentration; data are calculated ratios from panel A.

Fig. 5. Quantitative Western blotting. ATP synthase concentration in chromatophores preparation both unfractionated (A and B) and in fractions along the sucrose gradient (C). (A) Aliquots of β-subunit and denatured chromatophores were loaded on a SDS-PAGE (in a random order to avoid any possible positional bias) and transblotted on a nitrocellulose membrane. Digital images of the chemiluminescence intensity were acquired and analyzed as described in Materials and methods. (B) The chemiluminescence intensities of the β-subunit were plotted (full circles) and fitted by a straight line. By means of this straight line the chemiluminescence intensities of the chromatophore samples were converted in amounts of β-subunit (empty circles). The corresponding Bchl concentration is indicated by the top X-axis. (C) Western blotting of sucrose density fractions of the chromatophore preparation used for EM imaging. Volumes containing 200 pmol of Bchl were loaded for each samples. The ATP synthase concentration was obtained by comparison with a calibration curve of purified β-subunit as in panel B. The two curves are the best fit to the data of Gaussian functions. An estimate of the average ATP synthase/Bchl ratio in non-fractionated chromatophores from this preparation can be obtained by taking the ratio of the two areas defined by the Gaussian functions, which gives a value of 1 ATP synthase per 236 Bchl molecules.

Fig. 6. Probability of finding x ATP synthase molecules per vesicle according to Poisson's distribution in four different samples: unfractionated chromatophores (full circles) and sucrose density fractions at 30% (empty circles) and 39% (empty triangles). These probability distributions were calculated as described in the Results according to Eq. (4), and with the aid of a QuickBASIC software routine. They involved the size distributions shown in Fig. 1 (negative staining data) and in Fig. 3A and C.

Fig. 7. Efrapeptin titration of carotenoid signal and ATP synthesis elicited by three flashes (fired 500 ms apart) under phosphorylating condition. (A) Carotenoid signal traces of chromatophores in the presence of 20 μM ADP and 5 mM Pi, at 36.4 μM Bchl. (a) Uninhibited chromatophores; (b) 60 nM efrapeptine; (c) 200 nM efrapeptine; (d) 200 nM efrapeptine and 20 μM oligomycin. The flashes are marked by the arrows. The decay of trace (c) following the 3rd flash was fitted to a multiexponential function (continuous line) and used in the analysis shown in panel C. Only few traces are shown for clarity. (B) Traces are from panel A and additional measurements. The time scale origin has been set to 1.5 ms after the third flash. Only the time interval up to 300 ms is shown for clarity. The continuous lines are the biexponential best fit functions of the data up to 800 ms after the third flash, which were used in the analysis shown in panel C. The numbers are nM efrapeptin concentrations for each trace. (C) “Phosphorylating currents” as a function of time for the efrapeptin concentrations shown in panel B. Each trace has been obtained by subtracting from the “total ionic current” (traces in panel B) the “passive ionic current” (trace 200 nM efrapeptin in panel A) as a function of the carotenoid signal amplitude, as described in the text. (D) Normalized initial current (full circles) and normalized current integral (empty circles), i.e., the translocated charge, were calculated as described in the text from traces of panels A and B and additional measurements. Empty triangles: ATP yield after the third flash, directly measured with luciferine–luciferase. The maximal value of the ATP yield in the absence of inhibitor was 0.66 nmol ATP/μmol Bchl. The curve through the full circles is the best fit to the data of the function describing the reversible binding of an inhibitor to an enzyme. The best fit value of the binding constant was 1.3 nM. The curve through the empty symbols is arbitrary and has been drawn by hand.

Parameters used for estimating the average number of ATPase molecules per vesicle in the non-fractionated chromatophores and in the gradient fractions at 30% and 39% sucrose concentration