Mechanism of photoinduced electron transfer in photosystem II

ISSN 0006-3509, Biophysics, 2007, Vol. 52, No. 1, pp. 40–45. © Pleiades Publishing, Inc., 2007. Original Russian Text © I.B. Klenina, W.O. Feikema, P.Gast, M.G. Zvereva, I.I. Proskuryakov, 2007, published in Biofizika, 2007, Vol. 52, No. 1, pp. 57–62. a

b

b

a

a

CELL BIOPHYSICS

Mechanism of Photoinduced Electron Transfer in Photosystem II Reaction Centers I. B. Kleninaa, W. O. Feikemab, P. Gastb, M. G. Zverevaa, and I. I. Proskuryakova aInstitute

of Basic Biological Problems, Russian Academy of Sciences, Pushchino, Moscow Region, 142290 Russia of Biophysics, Huygens Laboratory, P.O. Box 9504, 2300 RA Leiden, Netherlands

bDepartment

Received November 15, 2005; in final form, May 12, 2006

Abstract—The shape of the EPR spectrum of the triplet state of photosystem II reaction centers with a singly reduced primary acceptor complex QAFe2+ was studied. It was shown that the spectroscopic properties do not significantly change when the relaxation of the primary acceptor is accelerated and when the magnetic interaction between the reduced quinone molecule QA and the nonheme iron ion Fe2+ is disrupted. This observation confirmed the earlier conclusion that the anisotropy of the quantum yield of the triplet state is the main cause of the anomalous shape of the EPR spectrum. A scheme of primary processes in photosystem II that is consistent with the observed properties of the EPR spectrum of the triplet state is discussed. DOI: 10.1134/S0006350907010083 Key words: photosystem II, reaction center, electron transfer, triplet state, electron paramagnetic resonance

INTRODUCTION

Here, P is the chlorophyll dimer; B is monomeric chlorophyll; Pheo is pheophytin; QA and QB are the primary and secondary electron acceptors (plastoquinones), respectively; and Fe2+ is the nonheme iron ion. However, such straightforward analogy with bacterial photosynthesis often contradicts experimental data.

The current knowledge of photochemical processes in photosystem II (PS II) reaction centers is far from complete. These processes are of special interest because, in particular, they lead to generation of a high positive potential sufficient to oxidize water molecules.

In this work, we study the properties of the triplet state of PS II RCs. Such studies have played a great role in understanding the mechanism of primary processes in bacterial RCs [5], where the triplet state is populated as a result of the recombination of a radical pair P+Pheo– (here P is the bacteriochlorophyll dimer molecules and Pheo is bacteriopheophytin). For this purpose, since the radical pair is initially generated in the singlet state, its lifetime should be sufficient for a singlet–triplet transition. This condition is met if electron transfer is blocked by (singly or doubly) reducing the primary acceptor QA or removing it. At low temperature, the triplet state is located at P. The state 3P in bacterial RCs is readily detected by EPR. The triplet state of PS II RCs under similar

At the present time, it is widely believed that PS II reaction centers (RCs) of higher plants and cyanobacteria are close analogues of the RCs of phototrophic purple bacteria [1]. Recent data on the structure of PS II RCs of thermophilic bacteria appear to support this opinion, at least, on the spatial arrangement of electron transfer cofactors [2–4]. If we follow this analogy, then the sequence of reactions of photoinduced electron transfer in PS II RCs should be the following: P*BPheoQAFe2+QB → P+B–PheoQAFe2+QB → → P+BPheo–QAFe2+QB → P+BPheoQ −A Fe2+QB → → P+BPheoQAFe2+Q −B .

(1) 40

PHOTOINDUCED ELECTRON TRANSFER IN PS II REACTION CENTERS

conditions is located at B [6]. Until recently, this state could be studied by EPR only in samples with the primary acceptor QA doubly reduced or absent. The 3B spectra under the conditions of single reduction of QA have been detected but recently [7] using time-resolved EPR. The shape of the published [7] spectra differs significantly from the shape of the EPR spectrum of the triplet state of photosynthetic RCs [8]. The causes of this difference were analyzed. This analysis rejected such possibilities as the overlap of the second triplet (e.g., from a light-harvesting antenna), the anisotropic relaxation of 3B, and the spin evolution of the three-spin system P+Pheo–(Q −A Fe2+), in which the third spin S = 1/2 belongs to the quinone Q −A involved in a strong magnetic interaction with Fe2+ (S = 2). A further analysis showed that the spin evolution of the radical pair is anisotropic (i.e., the frequency of its singlet–triplet transitions depends on the orientation in an external magnetic field), which is because of its short lifetime. Hence it was concluded [7] that the mechanism of electron transfer in PS II RCs differs from that shown in scheme (1). In particular, it was assumed that the state P+B– lives much longer than in bacterial RCs. Since this result is of fundamental significance for developing general concepts of primary photosynthetic processes, all the reasoning should be thoroughly checked. The most vulnerable point seems to be the assumption that the evolution of the three-spin system P+Pheo–(Q −A Fe2+) is not the cause of the distortion of the spectrum. In this work, this possibility was subjected to additional experimental testing. EXPERIMENTAL In the experiments, we used PS II oxygen-evolving complexes that were obtained according to a published procedure [9] from spinach and contained about 35 chlorophyll molecules per RC. Before use, the preparation was kept in liquid nitrogen in BTS400 buffer (20 mM Bis-Tris, pH 6.5, 20 mM MgCl2, 5 mM CaCl2, 10 mM MgSO4, 0.4 M sucrose, and 0.03% n-dodecyl-β-D-maltoside). The magnetic interaction between QA and Fe2+ was disrupted either by transfer of PS II complexes to glycine buffer (pH 11) with subsequent return to BTS400 or by 90-min incubation with 350 mM KCN at pH 8. Previously, both methods were successfully used for the same purposes in treating larger BBY particles ([10] and [11], respectively). For PS II oxygen-evolving complexes, BIOPHYSICS

Vol. 52

No. 1

2007

41

these methods were used for the first time and proved efficient. The disruption of the interaction in the iron–quinone complex was evidenced by disappearance of the low-temperature EPR signal of Q −A Fe2+ at g = 1.82 and simultaneous appearance of the free-radical signal of Q −A at g = 2.0045, as in the literature [10, 11]. Single reduction of the primary acceptor was performed photochemically by freezing a sample in an EPR spectrometer cavity from 260 K to about 100 K during continuous exposure to unfocused light from a 150-W halogen lamp through a heat filter. The light intensity in the region of the sample was 10 mW/cm2. To obtain the doubly reduced state of QA, the preparations were incubated for 40 min in the dark in the presence of 8 mM sodium dithionite and 100 μM benzyl viologen in an N2 atmosphere. To increase the amplitude of the EPR signal of Q −A Fe2+, 200 mM sodium formate was added to some samples [12]. This additive did not affect the shape of the EPR spectra. Capillaries for EPR measurements were filled in an N2 atmosphere in dim light and frozen in total darkness. Whether the reduction of the primary acceptor was single or double was checked by the EPR signal at g = 1.82 and by the appearance of the signal of the triplet state of PS II RCs, which could be detected by CW EPR (see also Feikema et al. [7]). RCs of Blastochloris viridis (previously Rhodopseudomonas viridis) were isolated according to [13] and photoreduced in the cavity at low temperature. To improve the freezing conditions, all samples were in 66% glycerol. The time-resolved EPR measurements were performed according to a published procedure [14]. A sample was excited by bursts of the second harmonic of a YAG laser (532 nm, 4 ns, repetition frequency 9.7 Hz), the energy of which was reduced to 0.3 mJ to slow down the accumulation of the products of irreversible photochemical processes in RCs. The signals from the mixer of a 3-cm microwave EPR spectrometer were passed through a broadband amplifier to a stroboscopic integrator and recorded in 0.5–1.8 μs after the burst. The output signal of the integrator entered the recording system and was recorded during continuous scanning of the spectrometer magnetic field. To increase sensitivity, the spectrum was accumulated (up to 800 times) in the computer memory. The time resolution of the instrument was about 50 ns. Sample temperature was stabilized with an Oxford Instruments cryostat (United Kingdom) calibrated additionally against a carbon resistor.

42

KLENINA et al.

RESULTS

Fig. 1. Time-resolved EPR spectra of the triplet states of PS II oxygen-evolving complexes at 20K: (a) singly reduced primary receptor; (b) doubly reduced primary receptor; and (c) calculated spectrum of the triplet state at D = +0.0286 cm–1, E = +0.0044 cm–1, and the spin polarization corresponding to the selective occupation of the sublevel T0 (see Discussion). The arrow at the left indicates the direction in which the microwave power is absorbed. The arrows at the top indicate the positions of canonic fields of the triplet spectrum; the sequence corresponds to the case D, E > 0. The measurement conditions are described in the text.

Fig. 2. Effect of temperature on the time-resolved EPR spectra of RCs of the phototrophic bacterium B. viridis with the reduced primary acceptor at T = (a) 6 and (b) 30 K. Magnetophotoselection was cancelled by summing the spectra obtained at two orthogonal polarizations of the exciting light [8].

Figure 1a presents a typical EPR spectrum of the triplet state that is reproducibly observed at T = 20 K in untreated preparations of PS II oxygen-evolving complexes with the singly reduced primary acceptor QA. For comparison, Fig. 1b shows a similar spectrum for the doubly reduced primary acceptor. Both spectra are described by the same set of zero-field splitting parameters (D = +0.0286 cm–1, E = +0.0044 cm–1), which is indicative of the same position of location of the triplet state—monomeric chlorophyll B. However, the shapes of the spectra differ significantly. The EPR spectrum for the doubly reduced primary acceptor is close to that expected theoretically (Fig. 1c), whereas the spectrum of singly reduced PS II oxygen-evolving complexes has practically no Z components and significantly reduced X components. Lest the shape of the spectrum in Fig. 1a should be altered by the magnetic interaction with the iron–quinone complex, we carried out the experiments described below. The magnetic interaction between Q −A Fe2+ and the components of the radical pair most noticeably affects the shape of the spectrum of the triplet state in RCs of the phototrophic bacterium B. viridis. This interaction clearly manifests itself upon “thawing” of the spin–lattice relaxation of the iron–quinone complex of bacterial RCs (about 10 K [15]). An increase in temperature from 6 to 30 K actually leads to an abrupt change in the spectrum (Fig. 2). No such temperature dependence is observed for PS II oxygen-evolving complexes (Fig. 3): over an even wider temperature range (6–50 K) the shape of the EPR spectrum of the triplet state of PS II remains virtually unchanged. The transition of the nonheme iron of RCs into the low-spin (S = 0) state significantly affects the magnetic properties of the iron–quinone complex. If the shape of the spectrum of the triplet is related to the evolution of the three-spin system, this should lead to a marked change in the spectral characteristics of 3B. The EPR spectra of the triplet state in correspondingly treated PS II oxygen-evolving complexes are presented in Fig. 4. Both procedures disrupting the magnetic interaction in the Q −A Fe2+ complex lead to the shape of the EPR spectra that is close to that observed for untreated preparations (Fig. 1a). BIOPHYSICS

Vol. 52

No. 1

2007


43

DISCUSSION In the general case, the shape of the EPR spectrum of the triplet state is determined by the energies of spin sublevels in the magnetic field and by the dynamics governing their population [16]. The energies of the sublevels (and the values of the resonance fields) are described by the zero-field splitting parameters (D, E). These parameters can be used with good reliability to identify a triplet-carrying molecule. The relative amplitudes of the spectral components depend on the populations of the sublevels, which, in their turn, are determined by the nature of the triplet molecule and the mechanism of generation of the triplet state. The shape of the EPR spectrum of the RC triplet state is best characterized for RCs of phototrophic bacteria. The triplet states of RCs are unique in being populated by recombination of radical pairs. As a result of this, strong magnetic fields entail selective population of the spin sublevel T0, the energy of which is independent of the magnetic induction [17]. The shape of the spectra with such spin polarization is quite easy to calculate with high accuracy (see, e.g., [8]). Any deviations from the theoretical shape are readily seen and require explanation. This is the case for the triplet state of RCs of the phototrophic bacterium B. viridis (Fig. 2; see also [18, 19]). The anomalous shape of the EPR spectrum in this case is explained using the theory of spin evolution in the three-spin system P+Pheo–(Q −A Fe2+) [19, 20]. A specific feature of B. viridis RCs is a significant magnetic interaction between the iron–quinone complex of the primary acceptor and the bacteriopheophytin molecule, which arises after single reduction of both electron carriers. The interaction intensity is estimated from the splitting of the EPR signal of reduced bacteriopheophytin (about 6 mT [21]). Note that, in the case of PS II RCs, reduced pheophytin experiences an interaction of similar intensity (about 5.6 mT [22]). This similarity of the interaction intensities suggests that, in PS II RCs, the main cause of the anomalous shape of the spectrum of the triplet state (Fig. 1a) can also be the evolution of the three-spin system P+Pheo–(Q −A Fe2+). Two hallmarks of the three-spin system as a precursor of the triplet state were noted [18–20]. One of them is the strong dependence of the shape of the resultant spectrum on temperature, and the other is the anisotropy of the phenomena observed. Both are related to the properties of the complex BIOPHYSICS

Vol. 52

No. 1

2007

Fig. 3. Time-resolved EPR spectra of PS II oxygen-evolving complexes with the singly reduced primary acceptor at T = (a) 6, (b) 30, and (c) 50 K.

Fig. 4. Time-resolved EPR spectra of PS II oxygenevolving complexes with the singly reduced primary acceptor after transition of the nonheme iron of RCs into the low-spin state by treatment (a) at pH 11 and (b) by 350 mM KCN. Different signal-to-noise ratios are caused by differences in concentration between the preparations.

P+Pheo–(Q −A Fe2+), namely, the sharp acceleration of the spin– lattice relaxation with an increase in temperature and its significant magnetic anisotropy. Comparison of the temperature dependences of the triplet states of RCs of B. viridis (Fig. 2) and PS II oxygen-evolving complexes (Fig. 3) does not confirm the existence of similar mechanisms that determine the

44

KLENINA et al.

shape of the EPR line in the two objects. The three-spin mechanism is most reliably excluded by demonstrating the absence of the effect of the magnetic anisotropy of the primary acceptor of PS II RCs on the shape of the triplet spectrum. It is known that the anisotropy practically vanishes after transition of the iron ion in the complex Q −A Fe2+ into the low-spin (S = 0) state [10, 11]. Figure 4 shows that the corresponding treatments do not lead to any marked change in the shape of the EPR spectrum. Consequently, the spin evolution in the three-spin system P+Pheo–(Q −A Fe2+) indeed does not significantly manifest itself in the population of the triplet state in PS II RCs. Thus, we obtained reliable additional evidence that the anomalous shape of the spectrum of the triplet state in PS II RCs with the singly reduced acceptor is related to the short lifetime of the radical pair, as proposed earlier [7]. Let us discuss this mechanism in more detail. The singlet–triplet transition in the radical pairs that precedes the population of the triplet state of RCs is caused by two factors, specifically, differences in g factors and hyperfine interactions of partners [23]. The transition frequency is also affected by the magnetic dipole–dipole and exchange interactions in the radical pairs. In an external magnetic field, all these interactions are to some extent anisotropic, which makes the frequency of singlet–triplet transitions dependent on the orientation of the radical pairs in the external magnetic field. If the lifetime of the radical pairs significantly exceeds the period of the lowest of these frequencies, then anisotropy slightly affects the properties of the triplet state arising after recombination of the radical pairs. Conversely, if the radical pairs are short-lived, the shape of the EPR spectrum of the triplet state is distorted. This distortion (anisotropy of the quantum yield of triplets) is determined by the fact that, in some radical pairs, the singlet–

triplet transition has no time to be completed and these radical pairs recombine in the singlet state. This phenomenon was first detected and quantitatively studied for RCs of phototrophic bacteria [24]. This is most probably also true for PS II RCs with the singly reduced primary acceptor. Note that an attempt to explain the anomalous shape of the EPR spectrum by the anisotropy of the quantum yield in this case leads to a significant contradiction. The optically measured lifetime of the radical pair—precursor of the triplet state in PS II—proves long enough (about 30 ns at 20 K [6]) for anisotropic effects to be suppressed. Feikema et al. [7] proposed an explanation eliminating this contradiction. It is based on the similarity of the absorption spectra of the Pheo and B molecules in their Q bands at low temperature [25, 26], which makes the states of the radical pairs P+B– and P+Pheo– optically indistinguishable. Optical methods record the total lifetime of these radical pairs, whereas the singlet–triplet transition occurs only in the state P+Pheo– since the strong exchange interaction in the radical pair P+B– “freezes” the spin evolution [23]. The charge state of the primary acceptor differently modulates the energy of the radical pairs P+B– and P+Pheo– and, thereby, changes the lifetimes of these radical pairs. Doubly reduced QA is protonated [12], the lifetime of the radical pair P+Pheo– increases, and the quantum yield of the triplet state becomes isotropic (Fig. 1b). The short lifetime of the radical pair P+Pheo– in singly reduced QA also explains the absence of the effect of the third spin belonging to Q −A Fe2+, which just has no time to manifest itself. Summarizing the results of this and previous [7] works, we can propose the following scheme of primary processes of electron transfer in PS II RCs at low temperature

P3BPheoQ −A Fe2+QB ← 3[P+B–]PheoQ −A Fe2+QB ← 3[P+BPheo–]Q −A Fe2+QB ↑ ↓

Reduction of Q

P1B*PheoQAFe2+QB → 1[P+B–]Pheo]QAFe2+QB → 1[P+BPheo–]Q −A Fe2+QB → P+BPheoQ −A Fe2+QB. Here, brackets refer to the state of the radical pairs, and the singlet–triplet transition in the radical pair [P+BPheo–] and the reverse electron transfer take place upon blocking the transfer to the primary acceptor by its reduction.

(2)

Although the direct electron transfer looks similar to scheme (1), the processes in PS II RCs differ significantly. First, the initial stage in electron transfer is the reaction of photoreduction, rather than photooxidation. Further, the lifetime of the radical pair BIOPHYSICS

Vol. 52

No. 1

2007


[P+B–] in the singlet (and, probably, also triplet) state at low temperature is on the order of tens of nanoseconds. And last, the recombination of the radical pair 3[P+B–] leads to direct population of the state 3B– without intermediate T–T energy transfer. The final conclusion on the validity of scheme (2) can be drawn after additional studies, probably, using time-resolved optical methods. It should be taken into account that the spectra of pigments of PS II RCs strongly overlap in the Qy absorption band. Studies in the Qx absorption bands of pigments may be promising. ACKNOWLEDGMENTS

45

9. P. J. van Leeuwen, M. C. Nieveen, E. J. van de Meent, et al., Photosynth. Res. 28, 149 (1991). 10. Y. Deligiannakis, C. Jegerschöld, and A. W. Rutherford, Chem. Phys. Lett. 270, 564 (1997). 11. Y. Sanakis, V. Petrouleas, and B. A. Diner, Biochemistry 33, 9922 (1994). 12. F. J. E. van Mieghem, W. Nitschke, P. Mathis, and A. W. Rutherford, Biochim. Biophys. Acta 977, 207 (1989). 13. H. J. den Blanken and A. J. Hoff, Biochim. Biophys. Acta 681, 365 (1982).

We thank W. van de Meer for help in preparing PS II complexes.

14. M. K. Bosch, I. I. Proskuryakov, P. Gast, and A. J. Hoff, J. Phys. Chem. 100, 2384 (1996).

This work was supported by the Russian Academy of Sciences (program “Molecular and Cell Biology”), the Netherlands Organization for Scientific Research NWO (grant no. 047009008), and the Russian Foundation for Basic Research (project no. 02-0448850).

15. C. A. Wraight, FEBS Lett. 93, 283 (1978).

REFERENCES 1. B. A. Diner and G. T. Babcock, in Oxygenic Photosynthesis: The Light Reactions. Advances in Photosynthesis, Ed. by D. R. Ort and C. F. Yocum (Kluwer Acad. Publ., Dordrecht, 1996), V. 4, pp. 213–247. 2. A. Zouni, H.-T. Witt, J. Kern, et al., Nature 409, 739 (2001). 3. N. Kamiya and J.-R. Shen, Proc. Natl. Acad. Sci. USA 100, 98 (2003). 4. K. N. Ferreira, T. M. Iverson, K. Maghlaoui, et al., Science 303, 1831 (2004).

16. S. P. McGlynn, T. Azumi, and M. Kinoshita., Molecular Spectroscopy of the Triplet State (Englewood Cliffs, N.J., Prentice-Hall, 1969; Mir, Moscow, 1972). 17. M. C. Thurnauer, J. J. Katz, and J. R. Norris, Proc. Natl. Acad. Sci. USA 72, 3270 (1975). 18. I. I. Proskuryakov, A. Ya. Shkuropatov, V. M. Voznyak, and V. A. Shuvalov, Biofizika 33, 877 (1988). 19. J. S. van den Brink, H. Manikowski, P. Gast, and A. J. Hoff, Biochim. Biophys. Acta 1185, 177 (1994). 20. P. J. Hore, D. A. Hunter, F. G. H. van Wijk, et al., Biochim. Biophys. Acta 936, 249 (1988). 21. R. C. Prince, D. M. Tiede, J. P. Thornber, and P. L. Dutton, Biochim. Biophys. Acta 462, 467 (1977). 22. V. V. Klimov, E. Dolan, E. R. Shaw, and B. Ke, Proc. Natl. Acad. Sci. USA 77, 7227 (1980).

5. A. J. Hoff, Light Emission by Plants and Bacteria, Ed. by Govindjee, J. Amesz, and D. C. Fork (Acad. Press, Orlando,1986).

23. A. L. Buchachenko, R. Z. Sagdeev, and K. M. Salikhov, Magnetic and Spin Effects in Chemical Reactions (Nauka, Novosibirsk, 1978) [In Russian].

6. F. van Mieghem, K. Brettel, B. Hillmann, et al., Biochemistry 34, 4798 (1995).

24. S. G. Boxer, C. E. D. Chidsey, and M. G. Roelofs, Proc. Natl. Acad. Sci. USA 79, 4632 (1982).

7. W. O. Feikema, P. Gast, I. B. Klenina, and I. I. Proskuryakov, Biochim. Biophys. Acta 1709, 105 (2005).

25. S. P. Årsköld, V. M. Masters, B. J. Prince, et al., J. Amer. Chem. Soc. 125, 13063 (2003).

8. I. V. Borovykh, I. I. Proskuryakov, I. B. Klenina, et al., J. Phys. Chem. B 104, 4222 (2000).

26. B. A. Diner, E. Schlodder, P. J. Nixon, et al., Biochemistry 40, 9265 (2001).

BIOPHYSICS

Vol. 52

No. 1

2007