Dynamics of energy transfer and trapping in the light-harvesting antenna of Rhodopseudomonas viridis

Mapping the ultrafast flow of harvested solar energy in living photosynthetic cells

Photosynthesis transfers energy efficiently through a series of antenna complexes to the reaction center where charge separation occurs. Energy transfer in vivo is primarily monitored by measuring fluorescence signals from the small fraction of excitations that fail to result in charge separation. Here, we use two-dimensional electronic spectroscopy to follow the entire energy transfer process in a thriving culture of the purple bacteria, Rhodobacter sphaeroides. By removing contributions from scattered light, we extract the dynamics of energy transfer through the dense network of antenna complexes and into the reaction center. Simulations demonstrate that these dynamics constrain the membrane organization into small pools of core antenna complexes that rapidly trap energy absorbed by surrounding peripheral antenna complexes. The rapid trapping and limited back transfer of these excitations lead to transfer efficiencies of 83% and a small functional light-harvesting unit.

During photosynthesis, energy is transferred from photosynthetic antenna to reaction centers via ultrafast energy transfer. Here the authors track energy transfer in photosynthetic bacteria using two-dimensional electronic spectroscopy and show that these transfer dynamics constrain antenna complex organization.

Energy trapping and detrapping in reaction center mutants from Rhodobacter sphaeroides

Time-resolved fluorescence of chromatophores isolated from strains of Rhodobacter sphaeroides containing light harvesting complex I (LHI) and reaction center (RC) (no light harvesting complex II) was measured at several temperatures between 295 K and 10 K. Measurements were performed to investigate energy trapping from LHI to the RC in RC mutants that have a P/P(+) midpoint potential either above or below wild-type (WT). Six different strains were investigated: WT + LHI, four mutants with altered RC P/P(+) midpoint potentials, and an LHI-only strain. In the mutants with the highest P/P(+) midpoint potentials, the electron transfer rate decreases significantly, and at low temperatures it is possible to directly observe energy transfer from LHI to the RC by detecting the fluorescence kinetics from both complexes. In all mutants, fluorescence kinetics are multiexponential. To explain this, RC + LHI fluorescence kinetics were analyzed using target analysis in which specific kinetic models were compared. The kinetics at all temperatures can be well described with a model which accounts for the energy transfer between LHI and the RC and also includes the relaxation of the charge separated state P(+)H(A)(-), created in the RC as a result of the primary charge separation.

Lifetimes of bacteriochlorophyll fluorescence in Rhodopseudomonas viridis and Heliobacterium chlorum at low temperatures

Fluorescence lifetimes of isolated membranes of Rhodopseudomonas viridis were measured in the temperature range of 77 K to 25 K. At room temperature, the main component of the fluorescence decay of bacteriochlorophyll (BChl) b had a time constant of 50 ps. In contrast to other purple bacteria, the emission at low temperature was spectrally homogeneous and showed essentially single lifetimes of 140 ps at 77 K and 180 ps at 25 K, with the primary electron donor in the oxidized state. Taking into account the relative fluorescence yields with open and closed reaction centers, we arrive at numbers of 125 ps and 215 ps, respectively, for open reaction centers. These numbers are significantly smaller than expected on the basis of measurements of the efficiency of charge separation, perhaps suggesting that the excitation decay in the absence of reaction centers is considerably faster at low temperature than at room temperature. At least four different spectral components with different lifetimes were observed at 25 K in the emission of Heliobacterium chlorum, a short-wavelength component of about 30 ps and three longer-wavelength components of about 100 ps, 300 ps, and 900 ps. This indicates a strong heterogeneity in the emitting pigment, BChl g-808. The component with the shortest lifetime does not appear to be affected by the redox state of the reaction center and might reflect energy transfer to BChl g species which are connected to the reaction center.

Escape probability and trapping mechanism in purple bacteria: revisited

Despite intensive research for decades, the trapping mechanism in the core complex of purple bacteria is still under discussion. In this article, it is attempted to derive a conceptionally simple model that is consistent with all basic experimental observations and that allows definite conclusions on the trapping mechanism. Some experimental data reported in the literature are conflicting or incomplete. Therefore we repeated two already published experiments like the time-resolved fluorescence decay in LH1-only purple bacteria Rhodospirillum rubrum and Rhodopseudomonas viridis chromatophores with open and closed (Q(A)(-)) reaction centers. Furthermore, we measured fluorescence excitation spectra for both species under the two redox-conditions. These data, all measured at room temperature, were analyzed by a target analysis based on a three-state model (antenna, primary donor, and radical pair). All states were allowed to react reversibly and their decay channels were taken into consideration. This leads to seven rate constants to be determined. It turns out that a unique set of numerical values of these rate constants can be found, when further experimental constraints are met simultaneously, i.e. the ratio of the fluorescence yields in the open and closed (Q(A)(-)) states F(m)/F(o) approximately 2 and the P(+)H(-)-recombination kinetics of 3-6 ns. The model allows to define and to quantify escape probabilities and the transfer equilibrium. We conclude that trapping in LH1-only purple bacteria is largely transfer-to-the-trap-limited. Furthermore, the model predicts properties of the reaction center (RC) in its native LH1-environment. Within the framework of our model, the predicted P(+)H(-)-recombination kinetics are nearly indistinguishable for a hypothetically isolated RC and an antenna-RC complex, which is in contrast to published experimental data for physically isolated RCs. Therefore RC preparations may display modified kinetic properties.

Projection map of the reaction center-light harvesting 1 complex from Rhodopseudomonas viridis at 10 A resolution

The photosynthetic reaction center-light harvesting 1 complex from Rhodopseudomonas viridis was purified and reconstituted into two-dimensional crystals. The single-layered crystalline sheets with lattice parameters a=b=133.3 A and gamma=120 degrees were investigated by electron cryo-microscopy and the projection map at 10 A resolution was calculated. The opening diameter of the light-harvesting ring of 72 A is sufficient to allow slight movement of the reaction center within the ring. Based on characteristic features observed in the projection map, the mechanism of energy transfer from the light-harvesting 1 complex to the reaction center was discussed.

Antenna organization in purple bacteria investigated by means of fluorescence induction curves

Fluorescence induction curves of purple bacteria (Rs. rubrum, Rps. viridis and Rb. capsulatus) were measured in the sub-millisecond time range employing a xenon flash technique. The induction curves of all three species displayed a sigmoidal shape. Analysis of the curves showed that none of the species examined had an antenna organization of a lake (i.e. unrestricted energy transfer between photosynthetic units). The apparent time constants of inter-unit exciton transfer were estimated to be approximately 24 ps in the case of LHC 1-containing species (Rs. rubrum and Rps. viridis) and 40 ps in the case of the LHC 2-containing species Rb. capsulatus. This result demonstrates that LHC 2 (B800-850) acts as a sort of insulator between photosynthetic units. Assuming a coordination number of 6 in the LHC 1-containing species the mean single step energy transfer time between adjacent LHC 1 can be estimated to be 4-5 ps. This is not perfectly compatible with the much faster Förster transfer rate of <1ps that follows from the minimal chromophore-chromophore distances estimated from digital image processing of micrographs from stained membranes. It thus may be concluded that the photosynthetic units (reaction center plus LHC 1) are loosely arranged in the photosynthetic membrane, like in the fluid-mosaic-membrane model, rather than in a hexagonally crystalline configuration.

Energy transfer in the inhomogeneously broadened core antenna of purple bacteria: a simultaneous fit of low-intensity picosecond absorption and fluorescence kinetics

The excited state decay kinetics of chromatophores of the purple photosynthetic bacterium Rhodospirillum rubrum have been recorded at 77 K using picosecond absorption difference spectroscopy under strict annihilation free conditions. The kinetics are shown to be strongly detection wavelength dependent. A simultaneous kinetic modeling of these experiments together with earlier fluorescence kinetics by numerical integration of the appropriate master equation is performed. This model, which accounts for the spectral inhomogeneity of the core light-harvesting antenna of photosynthetic purple bacteria, reveals three qualitatively distinct stages of excitation transfer with different time scales. At first a fast transfer to a local energy minimum takes place (approximately 1 ps). This is followed by a much slower transfer between different energy minima (10-30 ps). The third component corresponds to the excitation transfer to the reaction center, which depends on its state (60 and 200 ps for open and closed, respectively) and seems also to be the bottleneck in the overall trapping time. An acceptable correspondence between theoretical and experimental decay kinetics is achieved at 77 K and at room temperature by assuming that the width of the inhomogeneous broadening is 10-15 nm and the mean residence time of the excitation in the antenna lattice site is 2-3 ps.

Long-wavelength absorbing antenna pigments and heterogeneous absorption bands concentrate excitons and increase absorption cross section

The light-harvesting apparatus of photosynthetic organisms is highly optimized with respect to efficient collection of excitation energy from photons of different wavelengths and with respect to a high quantum yield of the primary photochemistry. In many cases the primary donor is not an energetic trap as it absorbs hypsochromically compared to the most red-shifted antenna pigment present (long-wavelength antenna). The possible reasons for this as well as for the spectral heterogeneity which is generally found in antenna systems is examined on a theoretical basis using the approach of thermal equilibration of the excitation energy. The calculations show that long-wavelength antenna pigments and heterogeneous absorption bands lead to a concentration of excitons and an increased effective absorption cross section. The theoretically predicted trapping times agree remarkably well with experimental data from several organisms. It is shown that the kinetics of the energy transfer from a long-wavelength antenna pigment to a hypsochromically absorbing primary donor does not represent a major kinetic limitation. The development of long-wavelength antenna and spectrally heterogeneous absorption bands means an evolutionary advantage based on the chromatic adaptation of photosynthetic organelles to spectrally filtered light caused by self-absorption.