Benchmark-Code to invite tenders for a Cluster

Our research methods are concentrated in the optical spectroscopy range, focusing mostly on time-resolved optical spectroscopic methods suitable for the characterization of energy and electron transfer processes and for characterizing the properties of self-organized pigment complexes. Thus techniques like e.g. femtosecond transient absorption, femto/-picosecond fluorescence, femtosecond photon echo, and also solid state CP-MAS-NMR are being used. Theoretical aspects of our work gain more importance recently and cover quantum-mechanical calculations, theoretical description of excitonic systems, molecular modelling and development of advanced kinetic modelling procedures.

Higher Plant Photosystems
This project focuses on the understanding of the structure/function relationships in photosynthetic antenna/reaction centersystems of oxygen-evolving organism on the basis of detailed structural, kinetic and spectral models. The aim is to understand, on a molecular basis, the effects of pigment-protein and pigment-pigment interaction like e.g. electron-phonon coupling, inhomogeneous broadening, exciton coupling, energy transfer and localization in the antenna structures as well as quenching and stress protection mechanisms located in the antennae and/or reaction centers. We furthermore aim at elucidatingdetails of the early ET reactions mechanisms in the reaction centers and the characterization and localization various redox intermediates. This project has been started in the previous reporting period (Müller, Niklas, Lubitz, Holzwarth, 2003, Biophys. J. 85, 3889; Prokhorenko and Holzwarth, 2000, J.Phys.Chem.B, 104, 11563)and is now reaching its full capacity. For several decades it had been assumed that the early ET processes in the photosystems of oxygenic photosynthesis are well understood, basically assuming similar reaction mechanisms as for bacterial RCs. Thus, single-branched ET mechanisms and primary electron donors being equivalent to the special pair of bacterial RCs had been assumed. However, a critical assessment of the underlying data indicated to us that many unproven assumptions and hypotheses were involved in these interpretations. For isolated RCs of PS II (the D1/D2-cyt-b₅₅₉-complex) we were able to show that in fact the accessory Chl was the primary electron donor, at least at low temperature. For PS I we obtained first indications in our data that the previously assumed P700 pigment pair was also not the primary donor. Furthermore for PS I indirect evidence from the observation of two phylloquinone oxidation lifetimes emerged from other groups which suggested that ET may be occurring in both quasi-symmetric cofactor branches.

A detailed study and interpretation of the ET processes in PS I and PS II intact core systems requires however an understanding of the global energy transfer dynamics from the antenna to the RC, i.e. whether the energy transfer, in particular the transfer step from the antenna to the RC, is rate-limiting in the system or whether the ET is the principal bottleneck for the trapping in the radical pairs. This question has been discussed controversially since more than 3 decades (see Renger and Holzwarth, 2005, for a review). We have studied the trapping kinetics carefully by femtosecond transient absorption as well as fluorescence kinetics for several PS I core particles, for PS II cores from T. elongatus and for PS II particles with large antennae from higher plants (BBY particles). Fig. 1 shows that the total trapping kineticis indeed extremelytrap-limited, i.e. limited by the ET processes, rather than the energy transfer to the RC, in all core particles. Only in BBY particles the trapping kinetics is in the intermediate range, i.e. neither trap-limited nor diffusion-limited.

Photosystem I
Optical studies on PS I are complicated by the fact that the ET processes have to be studied with a background of 90 antenna Chls, which requires extremely low excitation intensities and a high S/N ratio of the data in order to be able to study the ET processes undisturbed by annihilation effects. Fortunately essentially all of the excited state energy equilibration occurs on a time scale faster than the ET processes. In a time-resolved fluorescence study we were able to proof that the primary charge separation is reversible in PS I RCs, in contrast to many hypotheses in the literature. This observation played a key role in the following interpretation of the ET processes and the overall trapping. Reversible charge separation from the early radical pairs is well-known for bacterial RCs and for PS II, but had been excluded for PS I for a long time.

In our first femtosecond study on PS I cores we resolved for the first time i) the transient difference spectrum for thereaction center excited state, and ii) the formation and decay of the primary radical pair and its intermediate spectrum directly from measurements on open PS I reaction centers, iii) showed that three radical pair states were necessary to described the total kinetics, and iv) demonstrated that the primary charge separation takes only 6-9 ps, i.e. by a factor of three shorter than assumed previously.

Figure 1.Scaling of the total trapping time for PS I core, PS II core, and PS II BBY particles (with LHC II attached). The data show that for the core particles of PS I and PS II the overall kinetics is strongly trap-limited. For BBY particles with a large antenna the kinetics is in the intermediate range, i.e. neither entirely trap- nor diffusion-limited.

However, the nature of the three resolved radical pair states could not be assigned from these data. In collaboration with the group of Prof. Lubitz we thus studied PS I particles with mutations in the ET chains aiming at modifying the ET reactions. These mutations involved the HC B656 mutant which largely changes the redox potential of P700. Another mutant was the TV A739 mutant which breaks the hydrogen bond to the carbonyl-Chl of the P_A Chl, which in turn changes the redox potential of P700, but to a lesser extent. A third mutant was the WAA 679 mutant which removes a possible π-π-interaction at P700. Surprisingly none of these mutations influenced the rate of the primary ET step. Only for the HC B656 mutant a major influence on the ET kinetics was observed. It was, however, the rate of the secondary, and not the primary ET step that was reduced.

Theseresults led us to propose an entirely different ET pathway than assumed so far: ET starts from the accessory Chl with the A₀ Chl as the primary acceptor. Only in the second ET step is P700 oxidized, thus reducing the previously oxidized Chl_acc. The third ET step then reduces the phylloquinone(s). The data also strongly suggested, in contrast to previous interpretations, that the B-branch was active in ET. Whether the A-branch was also active could not be concluded unequivocally. This question was addressed subsequently using two additional mutants, i.e. YF mutations of the tyrosines providing hydrogen bonds to the A₀ Chls in the A- and B-branches.

Figure 2. Lifetime density maps of the femtosecond transient absorption data for Chlamydomonas PS I RC mutants with YF mutations near the A0 Chls in the A (left) or B (right) branches.

The femtosecond transient data for these mutants are given in Fig. 2 which shows that these mutations have distinct influence on the kinetics. Further analysis then proved that in PS I both cofactor branches are active in ET, as shown in Fig. 3. The branching ratio for the two branches appears to be about 3:2 in favour of the A-branch. The YF mutation on either side does not entirely block the electron transfer activity, but largely reduces the primary electron transfer step due to the shift to more negative redox potential of the A₀ Chl(s) when the tyrosine hydrogen bond is removed.

With electron transfer occurring in both branches and the additional reversibility of the first electron transfer step, the kinetic scheme is extremely complex, providing a challenge to both the quality of the data and the quality of the data analysis procedures. We are presently working on the kinetic analysis of these data in terms of a detailed rate constant model for the electron transfer steps.

Photosystem II
We were able to show recently for isolated D1-D2-cyt_b559 RC complexes at low temperatures that the primary electron transfer does not start from one of the chlorophylls in the pair, but rather from the “monomeric” accessory chlorophyll. Presumably the pheophytin (Pheo) is then the first electron acceptor, creating a monomer Chl⁺ Pheo^- primary radical pair, that only in a subsequent second electron transfer step then develops into the well-known P680⁺Pheo^- radical pair. However, no demonstration of this mechanism existed at room temperature nor in more intact PS II core particles when we started our experiments. Femtosecond transient absorption experiments at room temperature clearly showed that Pheo was reduced in the first electron transfer step and remained reduced throughout the two following radical pair intermediates that could be resolved (Figs. 4A and 4B, Holzwarth et al. 2006, PNAS). For PS II cores the situation is more complicated since the kinetic model also has to include the energy transfer steps from the antenna. We found that the antenna to RC energy transfer steps could be described satisfactorily by two antenna compartments which belong to the CP43 and CP47 antenna pigments.

Figure 3.Bi-branched electron transfer in PS I. The arrows and their numbering indicate the sequence of the electron transfer steps.

Figure 4.Mechanism and rate constants of the primary electron transfer steps at room temperature in isolated D1-D2-cyt_b559 complexes (A) and species-associated difference spectra (B) as obtained from the kinetic model. The corresponding data are given for intact PS II core particlesat room temperature in panels C and D. (taken from Holzwarth et al., PNAS, 2006)

An additional difference of the PS II core particles to the isolated RC involves the presence of the quinone electron acceptor in the intact system, thus allowing reduction of Q_A. Besides the energy transfer steps we were also able to resolve the electron transfer steps in the intact system by femtosecond transient absorption measurements and kinetic compartment modelling (Figs. 4C and D). It was thus possible, for the first time, to isolate the actual excited RC state spectroscopically in the intact PS II core complex. Our data showed that isolated RCs and the intact system have the same electron transfer rates for the first two electron transfer steps. This finding was in contrast to several previous reports from other groups which claimed that for the isolated D1-D2-cyt_b559 complex both the spectral properties and in particular the electron transfer rates would differ largely from the intact system. However, our data clearly show that isolated D1-D2-cyt_b559 complexes are good RC preparations whose electron transfer rates are not modified significantly from the intact system. Also spectroscopically at best a shift in the absorption spectrum of the D1-D2-cyt_b559 complex of about 1 nm would be compatible with our data. This is very encouraging and provides a solid basis for further detailed studies of the charge separation processes in PS II using isolated D1-D2-cyt_b559 complexes. Our data from time-resolved kinetics are in full agreement with the transient absorption data and yield the same kinetic model. We are now planning studies on PS II cores and isolated RCs with point mutations near the electron transfer cofactors. We are presently also studying the electron transfer processes in PS II cores with a reduced quinone Q_A acceptor (closed RCs). Our preliminary data indicate that there might occur a so-far unrecognized switching in the electron transfer mechanism upon reduction of Q_A. Such a switching could play an important role in the protection of PS II against photodamage in high light.

According to our findings these structures contain closely stacked chlorins based on a monomer basic unit. Several of these stacks combine at a higher level to single or (in the green sulphur bacterium Chlorobium tepidum) to double rod structures. More recently two different structures for chlorosomes have been proposed by other groups in the literature: The first one is a lamellar structure, proposed on the basis of electron microscopy data,that uses chlorin dimers as basic building blocks and does not contain any rod elements. A still further structure, containing rod-elements, but based also on a dimeric chlorin building block, has recently been proposed based on solid-state NMR (Egawa et al., 2007, PNAS 104, 790). Thus there exists at present a large disparity between these three structural models. One should note that solid-state NMR provides information on short-range order only, but can not provide any information on long-range order. We thus resorted to molecular modelling to solve that problem, as our spectroscopic (optical and NMR) data appear to be only in agreement with a monomeric stacking model for the chlorins. Electron microscopy, on the other hand, does not have the resolution to determine the short-range order, but should give information on the long-range order.

Figure 7.Top: AFM picture of the rod structures of the artificial BChl antenna aggregates. Bottom: Absorption spectra of the chlorin aggregates and the attached NBI chromophores (from Röger et al., 2006).

We note, however, that for an artificial chlorin aggregate it has been shown recently (for the first time) by atomic force microscopy (Röger et al., 2006)that slightly modified green bacterial chlorins indeed self-assemble as rod structures in vitro. Since their optical properties were very close to those of native chlorosomes this provides strong support for our chlorosomal model.In view of these discrepancies in the structures of chlorosomes proposed by different groups we are presently working on chlorosomes isolated from chlorosome mutants that simplify the heterogeneous chlorophyll side groups present in w.t. green bacterial chlorosomes, but do not change the optical properties. This should allow us to gain higher resolution in solid-state NMR and thus more insight into the structural arrangements.

Our activities in this reporting period were focused in four directions: i) Structural studies to improve our understanding of the molecular interactions that guide and control the self-aggregation of metallo-chlorins and the formation of supramolecular structures in natural systems and in artificial aggregates that may serve as aritificial self-organized antenna system in artificial photosynthesis, ii) the understanding of the role of syn- or anti-ligation of the metal centers in chlorins (Garcia-Martin et al. 2006; de Boer et al. 2004)which plays an important role in the chlorosomal structures, iii) the study of chlorin- and porphyrin-fullerene dyads as models for artificial electron transfer devices to be used in artificial photosynthesis (Schuster et al. 2004; Katterle et al. 2006), and iv) the development and characterization of efficient artificial self-assembling light-harvesting systems that have a very broad coverage of the solar spectrum, unlike pure chlorin systems, which have a large gap, the so-called “green gap” in their absorption properties (Röger et al. 2006) in collaboration with the group of Prof. F. Würthner, Würzburg.

All of these questions are important for the understanding of the structure/function relationships governing efficient light-harvesting properties in self-assembled systems and thus for the design of the artificial antenna and electron transfer units. In the following we discuss in more detail our strategy to build artificial antenna systems possessing a broad coverage of the solar spectrum and efficient light-harvesting.

In collaboration with the group of Prof. Würthner, Würzburg, we studied a green-absorbing naphthalinbisimide (NBI) chromophore covalently coupled to a chemically modified BChl-c analog. This coupling does not hinder the self-aggregation of the chlorin moiety to supramolecular rod structures as shown in Fig. 7 (top). This is also supported by the spectral properties of the chlorin part in comparison to the chlorin aggregate alone (Fig. 7 bottom). The energy transfer from the NBI chromophore to the chlorin aggregate was found to be more than 99% efficient, as shown by time-resolved fluorescence spectroscopy (Fig. 7). The coupling of the NBI chromophore increased the light-absorption efficiency for the solar spectrum by 26% as compared to a pure chlorin aggregate. Recently we have further increased the solar coverage substantially by covalently attaching a third chromophore to this system, thus basically closing the absorption gap (Röger et al., unpublished). Other approaches to build artificial antenna with a complete spectral coverage are also presently explored.

A key issue in these studies is also the detailed understanding of the excitonic properties of the supramolecular aggregates. We have carried out extensive theoretical studies which led to a better characterization of the exciton structure. A surprising result of these studies is the fact that the coherent excitonic state seems to be unusually long-lived in both chlorosomes and even more so in artificial aggregates, in the order of a few ps, and that the coherence is maintained at room temperature over a range of about 100 chlorin molecules. The theoretical exciton studies now provided an understanding of these unusually long-lived coherence states. Our studies suggest that strong exciton coupling reduces efficiently the exciton-phonon-coupling, which is mainly responsible for the optical dephasing and the localization of excitons on single chromophores. We are presently continuing these theoretical studies, in combination with experimental studies.

Carola S. Bösinger (2000) Regulations-mechanismen in Photosytem II: Picosekunden-aufgelöste Fluoreszenzuntersuchungen an isolierten Untereinheiten und ganzen Planzen. Heinrich-Heine-Universität Düsseldorf 

Dorte B. Steensgaard (2000) Structure- and Function Relationships in Chlorosomes, the Light-harvesting Complex in Green Photosynthetic Bacteria. University of Odense, Denmark

Martin Katterle (2001) Entwicklung von artifiziellen selbstorganisierenden Antennen-/ Elektronentransfer-Einheiten. Heinrich-Heine-Univ. Düsseldorf

European Union TMR project "Green Bacterial Photosynthesis" together with research groups in Denmark (Prof. M. Miller, Odense), Sweden (Prof. T. Gillbro, Umea), Spain (Dr. J. Garcia-Gil, Girona), The Netherlands (Prof. J. Amesz, Leiden). (1996- 2002).

Human Frontier Science Organization Award (HFSPO) on "Structure and function of supramolecular  bacteriochlorin assemblies" together with research groups in The Netherlands (Prof. H. de Groot, Chemistry Department Leiden University) and Japan (Prof. H. Tamiaki, Chemistry Department Ritsumeikan Univ.). (1997-2001);

European Union Demonstration Project in Biotechnology "High-Field solid state NMR structure determination on chlorosomes and artificial aggregates" within the Program "NMR structures of membrane proteins, complexes and lipid assemblies. A dedicated wide-bore ultra-high-field MAS NMR spectrometer for biological research" together with ten other multinational academic and industrial research groups (Main cooperation partner Prof. H. de Groot, Chemistry Department University of Leiden). (1997-2002),

Schwerpunkt-Programm "Intra- and Intermolecular electron transfer processes in Chemistry and Biology" Volkswagenstiftung (VW-Stiftung): "Self-Regulation Processes in biological intra- and inter-protein electron transfer" together with the group of Prof. V. Kharkyanen, Kiev, Ukraine (1998-2002).

Fonds der Chemischen Industrie (1998-2001)

EC Marie Curie Fellowship (Dr. S. Bernais-Barbry). Project on "Development of supramolecular self-assembled energy transfer/electron transfer devices " (2001-2002).

EC Marie Curie Research Training Network (RTN) INTRO2 for “Training and Research on Photosystem 2” (2004-2007) involving 9 European Research groups.

Co-Chair and Organizer European Brainstorming Workshop on "Solar Energy Conversion" (Regensburg, Germany, May 10-13, 2006). See report "Harnessing Solar Energy for the Production of Clean Fuels".

Chemistry Department, Arizona State University (ASU) Tempe, Arizona, USA, 2002, 2½ months (with group of Profs. D. Gust,  T. and A. Moore);  Studies of artificial self-assembled electron transfer units.

Chemistry Department, University of California Berkeley, Berkeley, California, USA, 2003, 3 months (with group of Prof. G. Fleming); Femtosecond photon echo peak shift studies on isolated chlorophyll pigments.