Photocycle of the flavin-binding photoreceptor AppA, a bacterial transcriptional antirepressor of photosynthesis genes

The flavoprotein AppA from Rhodobacter sphaeroides contains an N-terminal domain belonging to a new class of photoreceptors designated BLUF domains. AppA was shown to control photosynthesis gene expression in response to blue light and oxygen tension. We have investigated the photocycle of the AppA BLUF domain by ultrafast fluorescence, femtosecond transient absorption, and nanosecond flash-photolysis spectroscopy. Time-resolved fluorescence experiments revealed four components of flavin adenine dinucleotide (FAD) excited-state decay, with lifetimes of 25 ps, 150 ps, 670 ps, and 3.8 ns. Ultrafast transient absorption spectroscopy revealed rapid internal conversion and vibrational cooling processes on excited FAD with time constants of 250 fs and 1.2 ps, and a multiexponential decay with effective time constants of 90 ps, 590 ps, and 2.7 ns. Concomitant with the decay of excited FAD, the rise of a species with a narrow absorption difference band near 495 nm was detected which spectrally resembles the long-living signaling state of AppA. Consistent with these results, the nanosecond flash-photolysis measurements indicated that formation of the signaling state was complete within the time resolution of 10 ns. No further changes were detected up to 15 micros. The quantum yield of the signaling-state formation was determined to be 24%. Thus, the signaling state of the AppA BLUF domain is formed on the ultrafast time scale directly from the FAD singlet excited state, without any apparent intermediate, and remains stable over 12 decades of time. In parallel with the signaling state, the FAD triplet state is formed from the FAD singlet excited state at 9% efficiency as a side reaction of the AppA photocycle.     

Fluorescence spectral fluctuations of single LH2 complexes from Rhodopseudomonas acidophila strain 10050

We have investigated the energy landscape of the bacterial photosynthetic peripheral light-harvesting complex LH2 of purple bacterium Rhodopseudomonas acidophila by monitoring sequences of fluorescence spectra of single LH2 assemblies, at room temperature, with different excitation intensities as well as at elevated temperatures, utilizing a confocal microscope. The fluorescence peak wavelength of individual LH2 complexes was found to abruptly move between quasi-stable levels differing by up to 30 nm. These spectral shifts either to the blue or to the red were accompanied by a broadening and decrease of the intensity of the fluorescence spectrum. The frequency and size of these fluorescence peak movements were found to increase linearly with excitation intensity. Using the modified Redfield theory, changes in the realization of the static disorder accounted for the observed changes in spectral shape and intensity. Long lifetimes of the quasi-stable states suggest large free energy barriers between the different realizations.     

Stark effect measurements on monomers and trimers of reconstituted light-harvesting complex II of plants

The electric-field induced absorption changes (Stark effect) of reconstituted light-harvesting complex II (LHCII) in different oligomerisation states-monomers and trimers-with different xanthophyll content have been probed at 77 K. The Stark spectra of the reconstituted control samples, containing the xanthophylls lutein and neoxanthin, are very similar to previously reported spectra of native LHCII. Reconstituted LHCII, containing lutein but no neoxanthin, shows a similar electrooptical response in the Chl a region, but the Stark signal of Chl b around 650 nm amounts to at most approximately 25% of that of the control samples. We conclude that neoxanthin strongly modifies the electronic states of the nearby Chl b molecules causing a large electrooptical response at 650 nm stemming from one or more Chls b in the control samples. Ambiguities about the assignment of several bands in the Soret region [Biochim. Biophys. Acta 1605 (2003) 83] are resolved and the striking difference in electric field response between the two lutein molecules is confirmed. The Stark effect in the carotenoid spectral region in both control and neoxanthin-deficient samples is almost identical, showing that the neoxanthin Stark signal is small and much less intense than the lutein Stark signal.     

Photoisomerization and photoionization of the photoactive yellow protein chromophore in solution

Dispersed pump-dump-probe spectroscopy has the ability to characterize and identify the underlying ultrafast dynamical processes in complicated chemical and biological systems. This technique builds on traditional pump-probe techniques by exploring both ground- and excited-state dynamics and characterizing the connectivity between constituent transient states. We have used the dispersed pump-dump-probe technique to investigate the ground-state dynamics and competing excited-state processes in the excitation-induced ultrafast dynamics of thiomethyl p-coumaric acid, a model chromophore for the photoreceptor photoactive yellow protein. Our results demonstrate the parallel formation of two relaxation pathways (with multiple transient states) that jointly lead to two different types of photochemistry: cis-trans isomerization and detachment of a hydrated electron. The relative transition rates and quantum yields of both pathways have been determined. We find that the relaxation of the photoexcited chromophores involves multiple, transient ground-state intermediates and the chromophore in solution does not generate persistent photoisomerized products, but instead undergoes photoionization resulting in the generation of detached electrons and radicals. These results are of great value in interpreting the more complex dynamical changes in the optical properties of the photoactive yellow protein.     

Mixed Exciton-Charge-Transfer States in Photosystem II: Stark Spectroscopy on Site-Directed Mutants

Sickle erythrocytes exhibit abnormal morphology and membrane mechanics under deoxygenated conditions due to the polymerization of hemoglobin S. We employed dissipative particle dynamics to extend a validated multiscale model of red blood cells (RBCs) to represent different sickle cell morphologies based on a simulated annealing procedure and experimental observations. We quantified cell distortion using asphericity and elliptical shape factors, and the results were consistent with a medical image analysis. We then studied the rheology and dynamics of sickle RBC suspensions under constant shear and in a tube. In shear flow, the transition from shear-thinning to shear-independent flow revealed a profound effect of cell membrane stiffening during deoxygenation, with granular RBC shapes leading to the greatest viscosity. In tube flow, the increase of flow resistance by granular RBCs was also greater than the resistance of blood flow with sickle-shape RBCs. However, no occlusion was observed in a straight tube under any conditions unless an adhesive dynamics model was explicitly incorporated into simulations that partially trapped sickle RBCs, which led to full occlusion in some cases.     

Bacteriochlorophyll fluorescence of purple bacteria at low redox potentials. The relationship between reaction center triplet yield and the emission yield.

This work describes fluorescence yield measurements in suspensions of strains of Rhodospirillum rubrum and Rhodopseudomonas sphaeroides in which the iron . quinone complex (X) was chemically reduced (state [PIX-]; P is the reaction center bacteriochlorophyll dimer, I is the long wavelength bacteriopheophytin), and compares these with the fluorescence observed when all the traps are open (state [PIX]) and with the fluorescence observed when all the traps are closed (state [P+IX]). At 77 K the amplitude and the shape of the fluorescence emission spectrum in [PIX-] are identical to those observed in [PIX]. This is a strong indication that all the extra fluorescence observed at room temperature in [PIX-] is, in fact, caused by an efficient back reaction [P+I-X-] leads to [P*IX-]. Using an equation similar to the original Vredenberg-Duysens relationship (Vredenburg, W.J. and Duysens, L.N.M. (1963) Nature 197, 355-357) but now assuming that a single reaction center has a probability pt of trapping an excitation and (1--pt) of re-emitting it to the surroundings, we are able to calculate pt as a function of the temperature by measuring the fluorescence in [PIX], [PIX-] and [P+IX] as a function of the temperature. The calculated pt values agree reasonably well with triplet yields measured in isolated reaction centers. Finally, we have measured the reaction center triplet yield (PTR) in intact systems and we have shown that the sum of the triplet yield and the remaining loss processes (PL) in the antenna bacteriochlorophyll including the bacteriochlorophyll dimer (such as fluorescence, internal conversion or direct triplet formation) is approximately constant; if we assume that at 77 K the only process which occurs in the reaction center is the formation of a reaction center triplet, than PTR + PL=1. The energy barrier between [P*IX-] and [P+I-X-] was estimated to be 0.11--0.15 eV for a set of preparations.     

Binding Stoichiometry of the Gene 32 Protein of Phage T4 in the Complex with Single Stranded DNA Deduced from Boundary Sedimentation

Abstract

Short 145 base DNA fragments in complex with the helix destabilizing protein of bacteriophage T4, GP32, have been studied with boundary sedimentation. The sedimentation coefficient was determined as a function of concentration, protein-nucleic acid ratio, temperature and salt concentration. It can be concluded that the measured values reflect the properties of the saturated DNA-GP32 complex. A combination of the earlier obtained translational diffusion coefficient of the complex with the sedimentation coefficient yields its anhydrous molecular weight (M_w = 5.4 · 10^s D), which corresponds to a size of the binding site of 10 nucleotides per protein. This procedure is not sensitive to the presence of non-binding protein molecules and to the assumed protein concentration, and therefore, it seems more reliable than a determination from titration experiments.

Similar sedimentation measurements were performed with tRNA-complexes containing 76 nucleotides. The translational diffusion coefficient can be calculated from the measured rotational diffusion coefficient and assuming the same hydrodynamic diameter for this complex as obtained for the 145 b DNA complex. The molecular weight derived from the data then also leads to a binding site size of about 10 nucleotides. This suggests that also the short tRNA-complex forms an open, strongly solvated structure, as was proposed for the 145 b DNA-GP32 complex.