Báo cáo khóa học: The function of D1-H332 in Photosystem II electron transport studied by thermoluminescence and chlorophyll fluorescence in site-directed mutants of Synechocystis 6803

doi:10.1111/j.1432-1033.2004.04287.x

Eur. J. Biochem. 271, 3523–3532 (2004) (cid:1) FEBS 2004

The function of D1-H332 in Photosystem II electron transport studied by thermoluminescence and chlorophyll fluorescence in site-directed mutants of Synechocystis 6803

Yagut Allahverdiyeva1, Zsuzsanna Dea´ k1, Andra´ s Szila´ rd1, Bruce A. Diner2, Peter J. Nixon3 and Imre Vass1 1Institute of Plant Biology, Biological Research Center, Szeged, Hungary; 2CR&D, Experimental Station, E.I. du Pont de Nemours and Co., Wilmington, DE, USA; 3Department of Biological Sciences, Imperial College London, UK

This is most probably caused by a decrease of Em(S2/S1). Concomitantly, the rate of electron donation from Mn to Tyr-Z• during the S1 to S2 transition is slowed down, relative to the wild type, 350- and 60-fold in the D1-H332E and D1- H332D mutants, respectively, but remains essentially unaf- fected in D1-H332S. A further effect of the D1-H332E and D1-H332D mutations is the retardation of the QA to QB electron transfer step as an indirect consequence of the donor side modification. Our data show that although the His residue in the D1-332 position can be substituted by other metal binding residues for binding photo-oxidisable Mn it is required for controlling the functional redox energetics of the Mn cluster.

– and S2QB

11

Keywords : Photosystem II; D1-protein; His332 mutants; thermoluminescence; flash-induced chlorophyll fluores- cence.

The His332 residue of the D1 protein has been identified as the likely ligand of the catalytic Mn ions in the water oxi- dizing complex (Ferreira, K.N., Iverson, T.M., Maghlaoui, K., Barber, J. & Iwata, S. (2004) Science 303, 1831–1838). However, its function has not been fully clarified. Here we used thermoluminescence and flash-induced chlorophyll fluorescence measurements to characterize the effect of the D1-H333E, D1-H332D and D1-H332S mutations on the electron transport of Photosystem II in intact cells of the cyanobacterium Synechocystis 6803. Although the mutants are not photoautotrophic they all show flash-induced thermoluminescence and chlorophyll fluorescence, which – recombinations originate from the S2QA demonstrating that charge stabilization takes place in the water oxidizing complex. However, the conversion of S2 to higher S states is inhibited and the energetic stability of the – charge pair is increased by 75, 50 and 7 mV in the D1- S2QA H332D, D1-H332E and D1-H332S mutants, respectively.

2 Photosystem II (PSII) catalyses the light-driven oxidation of water, evolving one molecule of O2 and four protons per two H2O molecules oxidized. The site of water oxidation is situated on the lumenal side of the thylakoid membrane, containing four Mn ions and the redox active tyrosine Tyr- Z [1–5]. One Ca2+ and one Cl– ion are also required for catalytic activity and appear to be located in the vicinity of the Mn cluster [2]. Although the structure of PSII was recently determined by X-ray crystallography at 3.5–3.8 A˚ resolution [6–8], the question how the local protein environment of the (Mn)4-Ca-Cl-Tyr-Z complex controls the oxidation of water has not been clarified.

may ligate the assembled Mn cluster [9–16]. The mutations of D1-His190 are not photoautotrophic [10,12,17–20], and with few exceptions do not evolve oxygen [12]. Many mutations of D1-His337 are unable to assemble a functional Mn cluster [13], while others are photoautrophic [13,21], or nonphotoautrophic, but can evolve oxygen [13]. According to the recent X-ray crystallography data D1-H337 appears to be hydrogen bonded to an oxygen atom that bridges two Mn, whereas D1-H190 is located in the vicinity of Tyr-Z, relatively far from the Mn cluster [8]. These observations imply that although both D1-H190 and D1-H337 are important residues for water oxidation they are unlikely to be involved in direct Mn ligation.

From site-directed mutagenesis studies three lumenal side histidines (His190, His337 and His332) of the D1 protein have been shown to be important for oxygen evolution, and by applying various spectroscopy methods including EPR, ESEEM and FTIR it was shown that at least one of them

3

The only histidine that appears to be a direct Mn ligand based on the currently available X-ray data is D1-H332 [8]. Extensive directed mutagenesis in combination with phy- siological and biophysical characterization has already been performed to clarify the role of this residue in oxygen evolution. The replacement of D1-H332 with Gln, Ser, Asn, Asp, Glu, Lys, Leu, Arg, Tyr and Gly resulted in nonphotoautotrophic strains [15]. Only the Gln and Ser mutants evolve O2 at 12–15% the rate of wild type (WT*) cells, the others are inactive in oxygen evolution. The only known photoautotrophic mutant at this site is a photo- autotrophic suppressor of D1-H332G, which has no compensatory mutations elsewhere in the coding region of

Correspondence to I. Vass, Institute of Plant Biology, Biological Research Center, Temesvari krt. 62, Szeged 6726, Hungary. Fax: +36 62 433 434, E-mail: imre@nucleus.szbk.u-szeged.hu Abbreviations: Chl, chlorophyll; DCMU, 3-(3,4-dichlorophenyl)-1,1- dimethylurea; PSII, photosystem II; TL, thermoluminescence; WOC, water oxidation complex. (Received 19 March 2004, revised 23 June 2004, accepted 12 July 2004)

3524 Y. Allahverdiyeva et al. (Eur. J. Biochem. 271)

(cid:1) FEBS 2004

single turnover saturating flashes (1 Hz) at +5 (cid:2)C. After excitation of the samples, TL was detected at a heating rate of 20 (cid:2)CÆmin)1.

Flash-induced fluorescence relaxation kinetics

Flash-induced increases and the subsequent decay of chlorophyll fluorescence yield were measured by a double- modulation fluorometer (PSI Instruments, Brno) [23] in the 150 ls to 100 s time range. The sample concentration was 5 lg Chl per mL. Multicomponent deconvolution of the measured curves was done by using a fitting function with two exponential and one hyperbolic component as des- cribed earlier [24]:

the psbA gene ([21]). Apart from this one, in all mutants except Asp and Glu substantial fractions of PSII complexes lack photo-oxidizable Mn ions in vivo [13], showing that D1-H332 influences the assembly and/or stability of the Mn cluster. EPR studies demonstrated that in D1-H332E PSII preparations the quantum yield for oxidizing the S1 state is very low, and the temperature threshold for forming the S2 state is approximately 100 K higher than in WT* PSII preparations [15]. Furthermore, the D1-H332E PSII parti- cles are unable to advance beyond the Tyr-Z•S2 state, as shown by the accumulation of a narrow (cid:1)split(cid:2) EPR signal under multiple turnover conditions [15]. Recent ESEEM studies have further confirmed the close association of D1-H332 with the Mn cluster in the S2 state [16].

FðtÞ (cid:1) F0 ¼ A1 expð(cid:1)t=T1Þ

ð1Þ

þ A2 expð(cid:1)t=T2Þ þ A3=ð1 þ t=T3Þ

where F(t) is the variable fluorescence yield, F0 is the basic fluorescence level before the flash, A1–A3 are the ampli- tudes, T1–T3 are the time constants. The nonlinear corre- lation between the fluorescence yield and the redox state of QA was corrected for by using the Joliot model [25] with a value of 0.5 for the energy-transfer parameter between PS II units.

Results

Thermoluminescence characteristics

Although data from the available literature demonstrate the importance of D1-H332 in water oxidation the mech- anistic details of its function are not clear. A topic of particular interest is the energetics of charge stabilization at the donor side of PSII in relation to the impaired function of the Mn cluster. In this study we used thermoluminescence (TL) and flash-induced fluorescence decay measurements to characterize the effect of the D1-H332D, D1-H332E and D1-H332S mutations on PSII electron transport in intact Synechocystis 6803 cells, with particular emphasis on the effects of charge stabilization energetics in the Mn cluster. Our results show that the energetic stability of the S2 state is significantly increased in the mutants, and suggest an important role of the D1-H332 residue in controlling the redox energetics of the Mn cluster.

Materials and methods

4

Propagation of cells

In untreated Synechocystis cells, the main TL band (B band) – recombination [26,27]. After single arises from the S2QB flash excitation this component appears at 22 (cid:2)C in the WT* cells, accompanied by a smaller peak at 52 (cid:2)C (Fig. 1). Flash-induced thermoluminescence was also observed in the D1-H332D, D1-H332E and D1-H332S mutants. However, the position of the B band was modified. In the D1-H332D strain the B band was up-shifted to 47 (cid:2)C, but a small component remained at around 20 (cid:2)C. In the D1-H332E and D1-H332S mutants the B band appeared at 28 (cid:2)C and 43 (cid:2)C, respectively.

Synechocystis sp. PCC 6803 cells were propagated in BG-11 growth medium containing 5 mM glucose, in a rotary shaker at 30 (cid:2)C under a 5% (v/v) CO2-enriched atmo- sphere. The intensity of white light during growth was 60 lEÆm)2Æs)1. Cells in the exponential growth phase (A580 ¼ 0.8–1) were used. All mutants were constructed in the psbA-3 gene of Synechocystis sp. PCC 6803 as described previously [9]. The WT* used in this study is the TC35 strain, which was constructed in an identical way to the D1-H332 mutants, except that the transforming plasmid carried no site-directed mutation.

Tris treatment

0 (cid:2)C8

6 5

and resuspended in growth medium

In the presence of 3-(3,4-dichlorophenyl)-1,1-dimethyl- urea (DCMU), which blocks the QA to QB electron transfer, – the main TL band (Q band) arises from the S2QA recombination [26,28]. In WT* cells the Q band is observed at around 12 (cid:2)C, and upshifted to 24, 18 and 23 (cid:2)C in the D1-H332D, D1-H332E and D1-H332S mutants, respect- ively (Fig. 1; Table 1). The shoulder that appears at around is due to a phase transition caused by melting of the sample. The TL band that is observed at around +47 (cid:2)C in the presence of DCMU (C band) is assigned to the – recombination [29,30]. The peak position of the Tyr-D•QA C band, and consequently the energetics of charge recom- – charge pair is not affected bination of the Tyr-D•QA significantly by the mutations (Fig. 1; Table 1).

To partially remove the Mn ions of the water oxidizing complex, cells were incubated with 0.8 M Tris (pH 8.5) for 30 min in dim light, and then were centrifuged at 10 000 g . This for 10 min treatment resulted in the removal of about 20–30% of photo-oxidizable Mn, as indicated by thermoluminescence and flash-induced chlorophyll fluorescence measurements.

Thermoluminescence

7

When excited by single turnover flashes, the amplitude of the B band undergoes a period-four oscillation, reflecting the redox cycling of the S states [26,27]. In WT* cells the well-known oscillatory pattern of dark-adapted samples can be seen, in which the first maximum appears after two flashes (Fig. 2). Mutation of D1-H332 leads to the loss or modification of oscillation in all strains studied showing a

This was measured with a home-built apparatus as in [22] in filter paper containing 50 lg chlorophyll (Chl) . Dark- adapted samples were excited either by continuous white light (10 WÆm)2) for 30 s at various low temperatures, or by

The function of D1-H332 in Photosytem II (Eur. J. Biochem. 271) 3525

(cid:1) FEBS 2004

Fig. 2. Flash-induced oscillation of the B thermoluminescence band in D1-H332 mutants. Cells were excited by a series of single saturating flashes at +5 (cid:2)C and the thermoluminescence B band was measured. The TL intensity was plotted as a function of exciting flash number for WT* (A), D1-H332D (B), D1-H332E (C) and D1-H332S (D) strains.

significant influence on the advancement of the S-state cycle. In the D1-H332D mutant the intensity of the band at 47 (cid:2)C gradually increases up to four-to-five flashes followed by a small decline, whereas in the D1-H332E mutant the intensity saturates after six-to-eight flashes. These observa- tions show that the advancement of the S-state cycle is blocked beyond S2 and also that the quantum yield of S2-state formation is low in the D1-H332E and D1-H332D mutants relative to the WT* cells. Although the oscillation of D1-H323S is rather flat (note the different vertical scales on the figure), its flash pattern is similar to that seen in the WT* cells when placed on a constant background.

Fig. 1. Thermoluminescence characteristics of D1-H332 mutants. Cells were excited by 30 s continuous illumination at )20 (cid:2)C and thermo- luminescence was measured in the absence (solid lines) or in the presence (dotted lines) of 10 lM DCMU. The measurements were performed on WT* (A), D1-H332D (B), D1-H332E (C) and D1-H332S (D) cells.

–)

Table 1. Thermoluminescence peak temperatures of the D1-H332 mutants. The data were obtained from the thermoluminescence bands shown in Fig. 1.

Strain B-band –) (S2QB ((cid:2)C) Q-band –) (S2QA ((cid:2)C) C-band (Tyr-DoxQA ((cid:2)C)

Modifications of the PSII donor side may lead to a change in the temperature dependence of S2-state formation as was observed after Ca2+-depletion [31,32]. As shown in Fig. 3. the threshold temperature for the low temperature block of the S1 fi S2 transition as measured by the induction of the B band was shifted to higher temperatures in case of the D1-H332D and D1-H332E strains. This conclusion is in agreement with the earlier results obtained by EPR studies on the D1-H332E mutant showing that the temperature threshold of S2 state formation is up-shifted [15]. In contrast, the temperature dependence of the B band induction in the

22 47 28 43 12 24 18 23 52 51 51 47 WT* H332D H332E H332S

3526 Y. Allahverdiyeva et al. (Eur. J. Biochem. 271)

(cid:1) FEBS 2004

D1-H332S strain was almost invariant in the )70 to )10 (cid:2)C range, similar to that observed in the WT* cells.

Relaxation kinetics of flash-induced fluorescence

Fig. 3. Temperature dependence of the B thermoluminescence band intensity in D1-H332 mutants. Cells were excited by a single saturating flash at various temperatures and the thermoluminescence B band was measured. The TL intensity was plotted as a function of excitation temperature for WT* (A), D1-H332D (B), D1-H332E (C) and D1-H332S (D) strains.

9 9

middle phase was slowed down from 3 to 20 ms, with a concomitant increase of the relative amplitude from 22 to 51%. Slow down of the fast phase was also observed in the D1-H332D mutant. However, this effect – increase of the time constant from 480 to 680 ls – was less pronounced than in D1-H332E. Interestingly, the slow phase was significantly slowed down in D1-H332E and could not be kinetically resolved in the 100 s time window of the measurement in D1-H332D, and appeared as a constant background. Slower fluorescence relaxation in the D1-

In dark-adapted samples illumination with a single satur- –, which results in a rapid rise of ating flash forms QA variable fluorescence. The subsequent relaxation of fluores- – [33– cence yield reflects different reoxidation routes of QA 35]. In the WT* cells, relaxation of the fluorescence yield after the flash is dominated by a fast component ((cid:2) 480 ls, – by QB 69%), which is related to the reoxidation of QA (Fig. 4A; Table 1). The middle phase ((cid:2) 3 ms, 22%) arises – reoxidation in centers, which had an empty QB from QA site at the time of the flash and have to bind a PQ molecule from the PQ pool. Finally the slow phase ((cid:2) 8 s, 9%) – reflects QAQB reoxidation through equilibrium with –QB via a reverse reaction QA (Table 2). with the S2 state Of the mutations studied, D1-H332E caused the most prominent effect on the relaxation kinetics (Fig. 4A). The time constant of the fast phase increased from 480 to 910 ls, and its relative amplitude decreased from 69 to 46%. The

Fig. 4. Relaxation of flash-induced chlorophyll fluorescence yield in D1-H332 mutants. The measurements were performed after single flash excitation in WT* (j), D1-H332D (d), D1-H332E (m) and D1-H332S (.) cells, in the absence (A) and in the presence (B) of 10 lM DCMU.

The function of D1-H332 in Photosytem II (Eur. J. Biochem. 271) 3527

(cid:1) FEBS 2004

Table 2. Characteristics of chlorophyll fluorescence yield relaxation in the D1-H332 mutants. Fluorescence was excited by single turnover flashes and relaxation of the fluorescence yield was measured as in Figs 4 and 5. The traces obtained in the absence of DCMU were fitted by two exponential (for the fast and middle phase) and one hyperbolic component (for the slow phase).

Mutant strain Fast phase s/Amp (ms/%) Middle phase s/Amp (ms/%) Slow phase s/Amp (s/%)

8.2 ± 2.3/9.0 ± 0.1 53.1 ± 17/9.0 ± 0.2 –/3.0 ± 0.1 0.48 ± 0.07/69 ± 2 0.68 ± 0.08/68 ± 3 0.91 ± 0.01/46 ± 3 0.36 ± 0.04/70 ± 4 3.2 ± 0.8/22 ± 1 5.4 ± 1.3/23 ± 3 20.9 ± 1.9/51 ± 2 3.3 ± 0.8/18 ± 4 6.1 ± 0.7/12 ± 0.7

1717

H332E and D1-H332D mutants as compared to the WT* was also indicated in the earlier work of Chu et al. from the increased level of fluorescence after 50 ms of the exciting flash [13]. These authors, however, did not provide detailed kinetic analysis of the effect.

the slow decaying phase in the presence of DCMU gradually increases with increasing flash number. The relative amplitude of this phase reached saturation after 50 flashes at 85% (Fig. 5B) indicating that the same percentage of PSII can form the S2 state. Multiple flash experiments were also performed with the other two mutants. In the D1-H332D strain the slow phase was 85% after one flash and saturated at 95% after 5–10 flashes (Fig. 5B). Whereas in the D1-H332S strain the 83% slow phase observed after one flash reached saturation at 90% after 5–10 flashes (Fig. 5B). These estimations agree well with the result of Chu et al. who found that (cid:2) 100, 95 and 78% of PSII centers contained a photo-oxidizable Mn cluster, in intact cells of the D1-H332E, D1-H332D and D1-H332S mutants, respectively [13].

When PQ binding to the QB site is blocked by DCMU the – occurs via charge recombination with reoxidation of QA donor side components. In intact WT* cells the fluorescence relaxation is dominated by a hyperbolic component arising – with the S2 state of the water from the recombination of QA oxidizing complex (Fig. 4B). Fast decaying fluorescence can be observed even in the presence of DCMU, when the donor side is impaired by mutations [9,13,36] or inhibited by UV-B light [24]. This fast fluorescence decay is expected to – and donors, reflect charge recombination between QA which are less stable than the S2 state [9,13,24,36].

A characteristic effect of the mutations was the increased time constant of the slow phase of fluorescence relaxation. This effect is most obvious in the D1-H332D mutant, where the time constant was increased to 22 s from about 1.2 s measured in the WT* cells (Fig. 4B; Table 2). The effect is smaller in the D1-H332E mutant (9 s) and almost negligible in D1-H332S (1.6 s). The increased time constants of the slow phase measured in the presence of DCMU indicate 75 – charge pair in the and 50 meV stabilization of the S2QA D1-H332D and D1-H332E mutants, respectively, but only about 7 meV in D1-H332S (Table 3).

No addition WT* H332D H332E H332S With DCMU WT* WT* + Tris H332D H332E H332S 2.2 ± 0.2/2.0 ± 0.5 2.6 ± 0.7/20 ± 2 12.2 ± 2.8/15 ± 2 6.5 ± 0.6/78 ± 2 2.7 ± 0.5/17 ± 1 1.22 ± 0.06/98.0 ± 1.0 1.16 ± 0.14/80 ± 2 22.3 ± 4.5/85 ± 2 8.6 ± 3.5/22 ± 2 1.6 ± 0.09/83 ± 2

Discussion

The effect of D1-H332 on S-state turnovers

Analysis of the decay curve in the presence of DCMU 10 resulted in a fast phase of 2.2 ms (2%) and a slow phase 1.2 s (98%) in the WT* cells, as well as 2.6 ms (20%) and 1.2 s (80%) in the Tris-treated cells (Table 2). In D1-H332E a dominant fast phase was observed after a single flash, which represents almost 80% of the whole relaxation (Fig. 4B). Fast decaying components were also observed in the D1-H332D and D1-H332S mutants; however, their contribution to the whole relaxation was only 15–17%. These data are in good agreement with previous fluores- cence measurements, which were analyzed by assuming three exponential decay components [13]. The time con- stants obtained for the fast phase were 6.5, 12.2 and 2.7 ms for the D1-H332E, D1-H332D and D1-H332S stains, respectively.

Previous literature data [15] and the TL results shown above indicate that the majority of PSII centers in the D1-H332E strain contain photo-oxidizable Mn clusters and can form the S2 state. Therefore the small amplitude – of fluorescence decay, which can be assigned to S2QA recombination, may look surprising. However, the earlier reports [15] and the gradual induction of the B TL band during a flash sequence also indicated that the S2 state formation has a low yield in the mutant. To clarify this point fluorescence relaxation was measured after induction by repetitive flashes given at 1 s intervals. As shown in Fig. 5A,

The existence of single flash-induced TL signals demon- strate that charge recombination with characteristics of the involvement of the S-states takes place after replacing the D1-H332 residue with Asp, Glu and Ser. The gradual increase of the B band intensity in D1-H332E shows that S2 state formation occurs with a low yield and also that the conversion of S2 to higher S-states is completely blocked. In contrast, the small decline of TL intensity after four flashes in D1-H332D indicates that the S2 fi S3 conversion is not fully blocked and may occur with a low probability. In D1-H332S the clear but small amplitude TL oscillation

3528 Y. Allahverdiyeva et al. (Eur. J. Biochem. 271)

(cid:1) FEBS 2004

by the D1-H332E mutation [15]. Our TL data are consistent with this finding and show a similar effect also for the D1-H332D mutant. The upshifted temperature threshold for S2-state formation was correlated with Ca2+ require- ment in spinach PSII particles [31,32]. In addition, the loss of Ca2+ induces the shift of the B and Q bands to higher temperatures [37–40]. These results might indicate that the affinity of Ca2+ binding at or around the Mn cluster is decreased as a result of mutating the D1-H332 residue to Asp and Glu, but not to Ser. However, the recent crystal structure indicates that the catalytic site of water oxidation is made up by a cubane-like Mn3CaO4 cluster linked to a fourth Mn [8]. D1-H332, together with D1-E189, appears to ligate one Mn in the cubane-like structure and does not interact directly with Ca2+, which is placed in the opposite corner of the cube [8]. Therefore, the phenomena indicating modified Ca2+ binding might be induced indirectly by structural changes caused by the replacement of D1-H332. It has to be noted that the increased temperature threshold of S2 state formation is known to be induced also by acetate treatment of PSII, which is expected to disrupt the hydrogen bonding network around the Mn cluster [3,20,41]. Thus, the modified temperature dependence of S2-state formation might be related to the different capacity of the Asp, Glu and Ser residues to replace His as the hydrogen donor (see further discussion below).

The effect of D1-H332 on electron donation from Mn to Tyr-Z

– recombination in intact cells.

A characteristic feature of the studied mutants is the biphasic decay of chlorophyll fluorescence yield in the presence of DCMU. The time constant of the fast decaying fluorescence component, which can be assigned to the Tyr- – recombination [13,24], is 12, 6.5 and 2.7 ms in the Z•QA D1-H332D, D1-H332E and D1-H332S cells, respectively. This is similar to that observed in Tris-treated cells (Table 2), as well as to previously reported fluorescence decay times in Synechocystis 6803 cells, in which the functioning of the Mn cluster was inhibited either by site- directed mutagenesis [12,13] or by ultraviolet radiation [24]. – recombination However, the time constant of Tyr-Z•QA appears to be significantly longer at 80–160 ms, determined from optical measurements performed in isolated PSII preparations of Synechocystis 6803 [15,42]. Thus, it is possible that the redox properties of the PSII reaction center are modified by the isolation procedure, and we prefer to assume that the fast decaying fluorescence indeed reflects the process of Tyr-Z•QA

indicates that S-state turnovers can take place in a fraction of centers, which is consistent with the 10–15% oxygen evolution in this mutant.

11 allowing the formation of the S2QA

It has been shown previously by EPR measurements that the temperature threshold for S2-state formation is increased

In the case of D1-H332E, the slow phase of fluorescence – recombination, represents decay, which arises from S2QA only 15% of the total decay amplitude after a single flash. This phenomenon is consistent with the low yield of S2 state formation, which can be explained by slow electron donation from the Mn cluster in the S1 state to Tyr-Z•. In normally functioning PSII, electron donation from the Mn cluster to Tyr-Z• occurs with an 80–160 ls time – constant [43–46]. This is much faster than Tyr-Z•QA recombination (which occurrs in the ms time range) – charge separated state with a high yield. However, if electron donation from the – recombination Mn cluster is retarded then the Tyr-Z•QA

Fig. 5. Formation of the S2 state under repetitive flash conditions in the D1-H332 mutants. (A) Flash dependence of chlorophyll fluorescence yield relaxation in D1-H332E. Cells were illuminated by series of sat- urating flashes in the presence of 10 lM DCMU and fluorescence decay was measured after the last flash. The applied flash series con- tained 1 (j), 2 (d), 5 (m), 10 (.), 50 (h) and 100 (r) flashes fired at 10 ms intervals. (B) Fraction of PSII centers in the S2 state. The relative amplitude of the slow phase of fluorescence decay, representing the – state was plotted as a function of amount of PSII centers in the S2QA flash number. The fraction of centers in which the S2 state was formed after each flash was calculated according to Eqn (1). The data are shown for D1-H332E (d), as well as for D1-H332D (j) and D1-H332S (m). The solid lines represent the best fit calculated data.

The function of D1-H332 in Photosytem II (Eur. J. Biochem. 271) 3529

(cid:1) FEBS 2004

–))c

–))d

Table 3. Free energy changes induced by the D1-H332 mutations. The data were calculated from the time constants of the different fluorescence decay phases listed in Table 2.

Strain DG(QA « QB)a meV DG(S2 « Tyr-Z)b meV D(DG(S2QA meV D(DG(Tyr-ZQA meV

– and QA/QA

a The free energy difference between QB/QB – was calculated by using DG ¼ kTÆln(Keq), where Keq equilibrium constant of sharing the electron between QA and QB is given by the inverse of the relative amplitude of the slow phase of fluorescence decay measured in the absence of DCMU (expressed as fraction of 1). b The free energy difference between Tyr-Z•/Tyr-Z and S2/S1 was obtained from the time constants of the fast and slow phases of fluorescence decay measured in the presence of DCMU as DG ¼ kTÆln(sslow/sfast). c The change of – recombination in the mutants relative to the WT* was calculated from the time constants of the slow phase of the free energy of the S2QA fluorescence decay measured in the presence of DCMU, using D(DG) ¼ DGmutant – DGWT* ¼ kTÆln(sslow, mutant/sslow, WT*). By using the assumption that the mutations do not affect the redox potential of QA, these values give the redox potential change of the S2 state. d The – recombination in the mutants relative to the WT* was calculated from the time constants of the change of the free energy of the Tyr-Z•QA fast phase of fluorescence decay measured in the presence of DCMU using the D(DG) ¼ DGmutant – DGWT* ¼ kTÆln(sfast, mutant/sfast, WT*) expression. By using the assumption that the mutations do not affect the redox potential of QA, these values give the redox potential change of Tyr-Z.

which appears to be smaller than observed in isolated PSII particles [15].

can compete with the photo-oxidation of Mn and decrease the yield of charge stabilization according to the following scheme:

k2 Q(cid:1)

k1 Tyr-ZÆ (cid:1)

ð2Þ

S1 (cid:1)!

A

The efficiency of electron donation from Mn to Tyr-Z• is also retarded in the D1-H332D mutant. However, in this case the fraction of Mn that is photo-oxidized after each flash is 0.95 and s1 ¼ 640 ls. This value is also significantly larger than the 80–160 ls in intact PSII centers, but about 60-fold less than in the D1-H332E mutant. In D1-H332S the fraction of Mn that is photo-oxidized after each flash is 0.9 and s1 ¼ 300 ls, which is similar to that obtained in fully functional PSII.

The effect of D1-H332 on the energetics of charge stabilization at the acceptor side of PSII

– and QA/QA

12

The fraction of S1 that is oxidized to S2 after single flash excitation is determined by the two rate constants as k1/(k1 + k2), or by the two time constants as s2/(s1 + s2). Upon excitation with repetitive flashes the S2 state is gradually accumulated if the time delay between the flashes – recombination. is shorter than the time needed for S2QA – charge pairs The fraction of centers with stabilized S2QA can be estimated from the gradual induction of the slow phase of fluorescence decay in the presence of DCMU. From the analysis of the results shown in Fig. 5B we obtained 0.15 for the fraction of photo-oxidizable centers in D1-H332E after each flash. Using this value and s2 ¼ 6.5 ms of the fast fluorescence phase we estimate s1 ¼ 37 ms for the electron donation from the S1 state to Tyr-Z• (Table 2). Based on similar analysis Chu et al. concluded that electron donation from Mn to Tyr-Z• is slowed down dramatically in intact cells of D1-H332E [13]. Debus et al. have also estimated s1 ¼ 800 ms for the electron transfer from Mn to Tyr-Z• during the S1 to S2 transition in isolated PSII particles of D1-H332E based on the – recombi- assumption that the s2 value of the Tyr-Z•QA nation is 160 ms [15]. The difference between our estimates and those of Debus et al. arises from the s2 values, i.e. the – recombination time constants, as in both case Tyr-Z•QA the s1 is about about 5.5 times larger than s2. The substantially different s2 values should be due to the different experimental material (intact cells in our case, and highly purified PSII complexes in the work of Debus et al. – is confirmed [15]) as the fast recombination of Tyr-Z•QA by the measurements performed in mildly Tris-treated cells (Table 2). Despite the difference in the numerical value of s2 our results confirm the finding of Debus et al. [13,15] about the significant slow down induced by the D1-H332E mutation of Mn oxidation by Tyr-Z•, and gives a reliable estimate for the modified time constant in intact cells,

The fast phase of fluorescence relaxation curves measured in the absence of DCMU reflects the kinetics of the QA to QB electron transfer process. Interestingly, both the D1-H332E and the D1-H332D mutations lead to a marked slow down of this process. In the D1-H332E mutant the amplitude of the fast phase was also decreased from 69 to 46% indicating that a large fraction of the PSII centers did not contain bound QB in the dark. The slow decaying phase of fluorescence measured in the absence of DCMU reflects –QB « – via the QA the recombination of the S2 state with QB – equilibrium. Therefore, QAQB its relative amplitude is proportional to the equilibrium constant of sharing the electron between QA and QB [35] from which the free energy – can be estimated. This gap between QB/QB calculation resulted in 62 meV for the WT* (Table 3), which is in good agreement with the (cid:2) 70 meV obtained in higher plant thylakoids by fluorescence [47] and by thermoluminescence measurements [48], and demonstrates the applicability of the method. In order to apply the estimation for the mutants we have to take into account that a single flash forms the S2 state only in a fraction of centers as discussed above. After correcting the amplitude of the fluorescence slow phase by the fraction of photo-oxidizable centers obtained in the previous section we estimated the – as 60, 41 and free energy gap between QB/QB

– and QA/QA

WT* H332D H332E H332S 62 60 41 52 162 193 184 163 – 75 50 7 – 44 28 5

3530 Y. Allahverdiyeva et al. (Eur. J. Biochem. 271)

(cid:1) FEBS 2004

13

reaction center Chl (P680*) [53]. In contrast, when charge recombination is monitored by fluorescence decay it is dominated by nonradiative processes [54]. Thus, the apparent lack of correlation between the stability change – when reported by fluorescence and TL charac- of S2QA teristics probably indicates that the D1-H332S mutation modifies the balance between radiative and nonradiative recombination routes.

52 meV in the D1-H332D, D1-H332E and D1-H332S mutants, respectively (Table 3). The significant decrease of the free energy gap in the D1-H332E mutant relative to the WT* strain probably arises from the modification of the QB site as indicated by the increased time constant and increased amplitude of the middle phase of fluorescence decay pointing to a decreased binding affinity of PQ at the QB site. Our results demonstrate that the D1-H332E and D1-H332D mutations modify the QA-to-QB electron trans- fer step on the other side of the PSII complex. Due to the large distance between D1-H332 and the QB binding region, which is about 45 A˚ , this effect should be caused by a structural change of the PSII complex that is transmitted from the donor side to the acceptor side. It is important to note that different treatments that alter the donor side of PSII have previously been demonstrated to influence the properties of QA and/or QB [49–52].

Our data show that significant stabilization of the S2 state and of Tyr-Z occurs in D1-H332D and D1-H332E, but not in the D1-H332S mutant. The different extent of stabiliza- tion observed as a consequence of different substitutions could be related to the protonation state of the residues. It is possible that Asp and Glu easily deprotonate or are already deprotonated, and they can stabilize the S2 state because of the anionic charge, which effect does not occur when Ser is used for substitution. This could also explain that the stabilization of Tyr-Z, which is located farther away, has the same tendency but with smaller values than the stabilization of the S2 state.

The effect of D1-H332 on the energetics of charge stabilization at the donor side of PSII

– and S2QA

i.e.

– and Tyr-Z•QA

The data obtained for the stability change of the S2 state and Tyr-Z indicate that the free energy gap between Tyr-Z•/ Tyr-Z and S2/S1 is also affected by the mutations, which can be calculated directly from the ratio of the time constants of the fast and slow phases of fluorescence decay in the presence of DCMU. We obtained 162 meV for the WT* cells, which is somewhat larger than the previous literature data estimating about 100 meV for this free energy gap (recently reviewed in [3,4]). The difference between our value and those in the literature probably arises from the experimental material, intact cells in our case and isolated PSII preparations in the previous studies. In the mutants the estimated free energy gap between Tyr-Z•/Tyr- Z and S2/S1 is 193, 184 and 163 meV for the D1-H332D, D1-H332E and D1-H332S strains, respectively (Table 3).

The role of D1-H332 in water oxidation

14

– and Tyr-Z•QA

The fast and slow phases of fluorescence decay curves measured in the presence of DCMU arise from the Tyr- – recombination, respectively. Therefore the Z•QA time constants provide information on the respective charge recombination energetics. As summarized in Table 3, the D1-H332D, D1-H332E and D1-H332S mutants increase – charge pair by the stabilization free energy of the Tyr-Z•QA – charge pair by 75, 50 44, 28 and 5 meV and that of S2QA and 7 meV, respectively. As the extent of stabilization is – it can be excluded that different for S2QA –) the effect could be caused by the increase of Em(QA/QA –) alone. Therefore we have to assume that either Em(QA/QA is left unchanged and the Em(Tyr-Z•/Tyr-Z) and Em(S2/S1) values are decreased, or that Em of all the three redox components is changed to different extent. The effect of the mutations on the thermoluminescence bands of the the D1-H332E and D1-H332E strains, which reflect increased stabilization free energy by increased peak tem- perature [53], support the first alternative. As shown in – recombination) and the B Table 1, both the Q band (S2QA – recombination) were shifted to higher temper- band (S2QB – atures, but the peak temperature of the C band (Tyr-D•QA recombination) was not affected significantly. Because it is unlikely that the mutation of D1-H332 would induce a parallel shift of the Tyr-D and QA free energy levels without affecting the gap between them it is straightforward to –) nor Em(Tyr-DÆ/Tyr-D) are assume that neither Em(QA/QA modified by the mutations. Based on this assumption we – charge assign the stabilization of the S2QA pair to the decreased level of Em(S2/S1) and Em(Tyr-Z•/ Tyr-Z), respectively.

the upshift of

The present findings and previous observations show that after replacement of D1-H332 with potential metal binding ligands, such as Asp, Glu and Ser, PSII retains a partially functioning Mn cluster [10,13,15,16]. According to the recent crystal structure the Mn ion that is ligated by D1-H332 has another protein ligand from D1-E189 [8], which could explain how photo-oxidizable Mn is retained at the binding site in the absence of the histidine residue. Electron abstraction from the Mn cluster is coupled with proton transfer reactions, which requires a hydrogen bonding network that provides an exit route for the proton released during water oxidation. It has been proposed that the D1-H332E mutation might affect this network of hydrogen bonds [16]. According to the model of Hoganson and Babcock the electron transfer step from Mn to Tyr-Z is coupled with proton abstraction and the hydrogen bond between Tyr-Z and D1-H190 is involved in the proton release pathway [3,55–57]. However, in the recent structural model of Ferreria and colleagues, D1-E189 was suggested to link the water oxidation site with a proton channel [8]. Because D1-H332 was proposed to ligate the same Mn as D1-E189 [8] it could also be part of the hydrogen bonding network. In line with this idea, the better performance of the Ser replacement may be related to better substitution for

It is of note that in the D1-H332S mutant there is a mismatch between the results of fluorescence decay and – thermoluminescence regarding stabilization of the S2QA charge pair. While the small increase of the fluorescence time constant indicates only a minor stabilization relative to the WT*, the Q band of TL is comparable to that in the D1-H332E and D1-H332D strains. A possible explanation is that TL reports charge recombination that proceeds via radiative processes and leads to the repopulation of the excited state of the

The function of D1-H332 in Photosytem II (Eur. J. Biochem. 271) 3531

(cid:1) FEBS 2004

by light-induced fourier transform infrared difference spectro- scopy. Biochemistry 38, 10187–10195.

hydrogen bonding by Ser than by Asp or Glu. Indeed the O-H group of Ser can act as an efficient H-donor, whereas neither Asp nor Glu are H-bond donors in their deproto- nated state. Moreover, the only other D1-H332 mutant that shows partial oxygen evolving activity is Gln [10,13], which also has a side chain that can serve as hydrogen donor. Thus it appears that the hydrogen bonding capacity of D1-H332 might be important for maintaining the exit route for the protons released during water oxidation.

15. Debus, R.J., Campbell, K.A., Peloquin, J.M., Pham, D.P. & Britt, R.D. (2000) Histidine 332 of the D1 polypeptide modulates the magnetic and redox properties of the manganese cluster and tyro- sine YZ in photosystem II. Biochemistry 39, 470–478.

16. Debus, R.J., Campbell, K.A., Gregor, W., Li, Z.-L., Burnap, R. & Britt, R.D. (2001) Does histidine 332 of the D1 polypeptide ligate the manganese cluster in photosystem II? An electron spin echo envelope modulation study. Biochemistry 40, 3690–3699.

Acknowledgements

15

17. Nixon, P.J., Chisholm, D.A. & Diner, B.A. (1992) Isolation and functional analysis of random and site-directed mutants of In Plant Protein Engineering (Shewry, P.R. & Photosystem II. Gutteridge, S., eds), pp. 93–141. Cambridge University Press, Cambridge.

This work was supported by a grant from the Hungarian Granting Agency OTKA (T034321) and from the NRICGP/USDA (2001- 35318-11270) to B.A.D. We are grateful to La´ szlo´ Sass for helping with the analysis of the oxidized S2 fraction from the multiple flash fluorescence experiments. 18. Roffey, R.A., Kramer, D.M., Govindjee & Sayre, R.T. (1994) Lumenal side histidine mutations in the D1 protein of Photo- system II affect donor side electron transfer in Chlamydomonas reinhardtii. Biochim. Biophys. Acta 1185, 257–270.

References

19. Hays, A.M., Vassiliev, I.R., Golbeck, J.H. & Debus, R.J. (1998) Role of D1-His190 in proton-coupled electron transfer reactions in photosystem II: a chemical complementation study. Biochem- istry 37, 11352–11365. 1. Andersson, B. & Styring, S. (1991) Photosystem II: Molecular organization, function, and acclimation. Curr. Top. Bioenerg. 16, 1–81.

20. Hays, A.M., Vassiliev, I.R., Golbeck, J.H. & Debus, R.J. (1999) Role of D1-His190 in the proton-coupled oxidation of tyrosine YZ in manganese-depleted photosystem II. Biochemistry 38, 11851– 11865.

2. Debus, R.J. (1992) The manganese and calcium ions of photo- synthetic oxygen evolution. Biochim. Biophys. Acta 1102, 269–352. 3. Tommos, C. & Babcock, G.T. (2000) Proton and hydrogen cur- rents in photosynthetic water oxidation. Biochim. Biophys. Acta 1458, 199–219.

21. Diner, B.A. (2001) Amino acid residues involved in the coordi- nation and assembly of the manganese cluster of photosystem II. Proton-coupled electron transport of the redox-active tyrosines and its relationship to water oxidation. Biochim. Biophys. Acta 1503, 147–163. 4. Renger, G. (2001) Photosynthetic water oxidation to molecular oxygen: apparatus and mechanism. Biochim. Biophys. Acta 1503, 210–228.

5. Nugent, J.H.A., Rich, A.M. & Evans, C.W. (2001) Photosynthetic water oxidation: towards a mechanism. Biochim. Biophys. Acta 1503, 138–146.

22. Vass, I., Cook, K.M., Dea´ k, Zs., Mayes, S.R. & Barber, J. (1992) Thermoluminescence and flash-oxygen characterization of the IC2 deletion mutant of Synechocystis sp. PCC 6803 lacking the photosystem II, 33 kDa protein. Biochim. Biophys. Acta 1102, 195–201.

6. Zouni, A., Witt, H.T., Kern, J., Fromme, P., Kraus, N., Saenger, W. & Orth, P. (2001) Crystal structure of photosystem II from Synechococcus elongatus at 3.8 A˚ resolution. Nature 409, 739–743.

7. Kamiya, N. & Shen, J.-R. (2003) Crystal structure of oxygen- evolving photosystem II from Thermosynechococcus vulcanus at 3.7 A˚ resolution. Proc. Natl Acad. Sci. USA 100, 98–103. 23. Trtilek, M., Kramer, D.M. & Koblizek, M. (1997) Dual-mod- ulation LED kinetic fluorometer. J. Lumin. 72–74, 597–599. 24. Vass, I., Kirilovsky, D. & Etienne, A.-L. (1999) UV-B radiation- induced donor- and acceptor-side modifications of Photosystem II in the cyanobacterium Synechocystis sp. PCC 6803. Biochemistry 38, 12786–12794.

8. Ferreira, K.N., Iverson, T.M., Maghlaoui, K., Barber, J. & Iwata, S. (2004) Architecture of the photosynthetic oxygen-evolving center. Science 303, 1831–1838. 25. Joliot, A. & Joliot, P. (1964) Etude cine´ tique de la re´ action pho- tochimique libe´ rant l’oxyge´ ne au cours de la photosynthe´ se. C. R. Acad. Sci. 258, 4622–4625.

9. Nixon, P.J. & Diner, B.A. (1992) Aspartate 170 of the Photo- system II reaction center polypeptide D1 is involved in the assembly of the oxygen-evolving manganese cluster. Biochemistry 31, 942–948. 26. Rutherford, A.W., Crofts, A.R. & Inoue, Y. (1982) Thermo- luminescence as a probe of photosystem II photochemistry: the origin of the flash-induced glow peaks. Biochim. Biophys. Acta 682, 457–465. 27. Demeter, S. & Vass, I. 10. Nixon, P.J. & Diner, B.A. (1994) Analysis of water-oxidation mutants constructed in the cyanobacterium Synechocystis sp. PCC 6803. Biochem. Soc. Trans. 22, 338–343.

(1984) Charge accumulation and recombination in photosystem II studied by thermoluminescence. I. Participation of the primary acceptor Q and secondary acceptor B in the generation of thermoluminescence of chloroplasts. Bio- chim. Biophys. Acta 764, 24–32. 11. Tang, X.S., Diner, B.A., Larsen, B.S., Gilchrist, M.L., Lorigan, G.A. & Britt, R.D. (1994) Identification of histidine at the catalytic site of the photosynthetic oxygen-evolving complex. Proc. Natl Acad. Sci. USA 91, 704–708.

28. Demeter, S., Herczeg, T., Droppa, M. & Horva´ th, G. (1979) Thermoluminescence characteristics of granal and agranal chloro- plasts of maize. FEBS Lett. 100, 321–324.

12. Chu, H.-A., Nguyen, A.P. & Debus, R.J. (1995) Amino acid residues that influence the binding of manganese or calcium to photosystem II. 1. The lumenal interhelical domains of the D1 polypeptide. Biochemistry 34, 5839–5858.

29. Demeter, S., Goussias, Ch, Bernat, G., Kovacs, L. & Petroluleas, V. (1993) Participation of the g¼1.9 and g¼1.82 EPR forms of the A Fe2+ of photosystem II in the semiquinone-iron complex, Q (cid:1) generation of the Q and C thermoluminescence bands, respect- ively. FEBS Lett. 336, 352–356. 13. Chu, H.-A., Nguyen, A.P. & Debus, R.J. (1995) Amino acid residues that influence the binding of manganese or calcium to photosystem II. 2. The carboxy-terminal domain of the D1 polypeptide. Biochemistry 34, 5859–5882.

30. Johnson, G.N., Boussac, A. & Rutherford, A.W. (1994) The ori- gin of 40–50 (cid:2)C thermoluminescence bands in Photosystem II. Biochim. Biophys. Acta 1184, 85–92. 14. Noguchi, T., Inoue, Y. & Tang, X.S. (1999) Structure of a histidine ligand in the photosynthetic oxygen-evolving complex as studied

3532 Y. Allahverdiyeva et al. (Eur. J. Biochem. 271)

(cid:1) FEBS 2004

31. Ono, T.A. & Inoue, Y. (1990) A marked upshift in threshold temperature for the S1-to S2 transition induced by low pH treatment of PSII membranes. Biochim. Biophys. Acta 1015, 373– 377. 44. Rappaport, F., Blanchard-Desce, M. & Lavergne, J. (1994) Kinetics of electron transfer and electrochromic change during the redox transitions of the photosynthetic oxygen-evolving complex. Biochim. Biophys. Acta 1184, 178–192.

45. Razeghifard, M.R. & Pace, R.J. (1997) Electron paramagnetic resonance kinetic studies of the S states in spinach PSII mem- branes. Biochim. Biophys. Acta 1322, 141–150.

32. Ono, T.-A., Kusunoki, M., Matsushita, T., Oyanagi, H. & Inoue, the (1991) Structural and functional modifications of Y. manganese cluster in Ca2+-depleted S1 and S2 states: Electron paramagnetic resonance and X-ray absorption spectroscopy studies. Biochemistry 30, 6836–6841. 33. Crofts, A.R. & Wraight, C.A. (1983) The electrochemical domain 46. Razeghifard, M.R., Klughammer, C. & Pace, R.J. (1997) Electron paramagnetic resonance kinetic studies of the S states in spinach thylakoids. Biochemistry 36, 86–92. of photosynthesis. Biochim. Biophys. Acta 726, 149–185.

47. Robinson, H.H. & Crofts, A. (1983) Kinetics of the oxidation- reduction reactions of the photosystem II quinone acceptor complex, and the pathway for deactivation. FEBS Lett. 153, 221– 226. 34. Crofts, A., Baroli, I., Kramer, D.M. & Taoka, S. (1993) Kinetics of electron transfer between QA and QB in wild type and herbicide- resistant mutants of Chlamydomonas reinhardtii. Z. Naturforsch. 48c, 259–266.

48. Demeter, S., Vass, I., Hideg, E´ . & Sallai, A. (1985) Comparative thermoluminescence study of triazine-resistant and – susceptible biotypes of Erigeron canadensis L. Biochim. Biophys. Acta 806, 16–27.

35. Renger, G., Eckert, H.-J., Bergmann, A., Bernarding, J., Liu, B., Napiwotzki, A., Reifarth, F. & Eichler, H.-J. (1995) Fluorescence and spectroscopic studies of exciton trapping and electron transfer in Photosystem II of higher plants. Aust. J. Plant Physiol. 22, 167– 181. 49. Vass, I., Ono, T.A. & Inoue, Y. (1987) Removal of 33 kDa protein charge pair in Photosystem II. sfecifically stabilizes the S2Q (cid:1) A FEBS Lett. 211, 215–220.

50. Krieger, A. & Weis, E. (1992) Energy-dependent quenching of chlorophyll-a-fluorescence: The involvement of proton-calcium exchange at photosystem II. Photosynthetica 27, 89–98. 36. Boerner, R.J., Nguyen, A.P., Barry, B.A. & Debus, R.J. (1992) Evidence from directed mutagenesis that aspartate 170 of the D1 polypeptide influence the assembly and/or stability of the man- ganese cluster in the photosynthetic water-splitting complex. Biochemistry 31, 6660–6672.

37. Ono, T. & Inoue, Y. (1989) Removal of Ca by pH 3.0 treatment inhibits S2 to S3 transition in photosynthetic oxygen evolution system. Biochim. Biophys. Acta 973, 443–449. 51. Krieger, A., Weis, E. & Demeter, S. (1993) Low-pH-induced Ca2+ ion release in the water-splitting system as accompanied by a shift in the midpoint redox potential of the primary quinone acceptor QA. Biochim. Biophys. Acta 1144, 411–418.

38. Ono, T., Izawa, S. & Inoue, Y. (1992) Structural and functional modulation of the manganese cluster in Ca2+-depleted photo- system II induced by binding of the 24-kilodalton extrinsic protein. Biochemistry 31, 7648–7655. 52. Johnson, G.N., Rutherford, A.W. & Krieger, A. (1995) A change in the midpoint potential of the quinone QA in Photosystem II associated with photoactivation of oxygen evolution. Biochim. Biophys. Acta 1229, 202–207.

16 39. Krieger, A., Weis, E. & Demeter, S. (1993) Low-pH-induced Ca2+ ion release in the water-splitting system is accompained with a shift in the midpoint redox potential of the primary quinone acceptor QA. Biochim. Biophys. Acta 1144, 411–418 . 53. Vass, I., Horvath, G., Herczeg, T. & Demeter, S. (1981) Photo- synthetic energy conservation investigated by thermolumine- scence. Activation energies and half-lives of thermoluminescence bands of chloroplasts determined by mathematical resolution of glow curves. Biochim. Biophys. Acta 634, 140–152. 40. Krieger, A., Rutherford, A.W. & Jegerscho¨ ld, C.

(1998) Thermoluminescence measurements on chloride-depleted and calcium-depleted photosystem II. Biochim. Biophys. Acta 1364, 46–54. 54. Rappaport, F., Guergova-Kuras, M., Nixon, P.J., Diner, B.A. & Lavergne, J. (2002) Kinetics and pathways of charge recombina- tion in photosystem II. Biochemistry 41, 8508–8517.

41. Haumann, M. & Junge, W. (1999) Photosynthetic water oxida- tion: a simplex-scheme of its partial reactions. Biochim. Biophys. Acta 1411, 86–91.

55. Hoganson, C.W., Lydakis-Simantiris, N., Tang, X.S., Tommos, C., Warncke, K., Babcock, G.T., Diner, B.A., McCracken, J. & Styring, S. (1995) A hydrogen-atom abstraction model for the function of YZ. in photosynthetic oxygen evolution. Photosynth. Res. 46, 177–184.

42. Metz, J.G., Nixon, P.J., Rogner, M., Brudvig, G.W. & Diner, B.A. (1989) Directed alteration of the D1 polypeptide of Photo- system II: Evidence that tyrosine-161 is the redox component, Z. connecting the oxygen-evolving complex to the primary electron donor, P680. Biochemistry 28, 6960–6969. 56. Hoganson, C.W. & Babcock, G.T. (1997) A metalloradical mechanism for the generation of oxygen from water in photo- synthesis. Science 277, 1953–1956.

57. Westphall, K.L., Tommos, C., Cukier, R.I. & Babrock, G.T. (2000) Concerted hydrogen-atom abstraction in photosynthetic water oxidation. Curr. Opin. Plant. Biol. 3, 236–242. 43. van Leeuwen, P.J., Heimann, C., Gast, P., Dekker, J.P. & van Gorkom, H.J. (1993) Flash-induced redox changes in oxygen- evolving spinach Photosystem II core particles. Photosynth. Res. 38, 169–176.