Effects of Ca2+ and EGTA on P680+ reduction kinetics and O2 evolution of Photosystem II

Biochimica et Biophysica Acta 1605 (2003) 21 – 34 www.bba-direct.com

S

Effects of Ca2+ and EGTA on P680 + reduction kinetics and O2 evolution of Photosystem II Gregory B. Stevens a,*, Philip B. Lukins b,c a

CSIRO Telecommunications and Industrial Physics, Bradfield Road, Lindfield, NSW 2070, Australia b Biophysical Optics Group, School of Physics, University of Sydney, NSW 2006, Australia c National Measurement Laboratory, Bradfield Road, Lindfield, NSW 2070, Australia Received 26 November 2002; received in revised form 15 May 2003; accepted 22 May 2003

Abstract

S

We report for the first time significant changes in the P680 + reduction kinetics of Photosystem II (PS II) in which the 17 and 23 kDa extrinsic polypeptides are intact, in the presence of Ca2 + or ethylene glycol bis (h-aminoethyl ether)-N,N,NV,NV-tetraacetic acid (EGTA) which were added to vary the Ca2 + concentration from 5 AM to 30 mM. The decrease in the extent of normal P680 + reduction decay with lifetimes of 40 – 370 ns and a corresponding increase in the extent of kinetics with lifetimes of 20 – 220 As was interpreted as being due to electron transfer from YZ to P680 + being replaced by slow forward conduction and by processes including P680 +/QA
recombination. The question of whether changes in P680 + reduction kinetics were caused by loss of Ca2 + from PS II or by direct interaction of EGTA with PS II was addressed by lowering the free-Ca2 + concentration of suspensions of PS II core complexes by serial dilution in the absence of EGTA. Despite a significant decrease in the rate of O2 evolution after this treatment, only small changes in the P680 + reduction kinetics were observed. Loss of Ca2 + did not affect P680 + reduction associated with electron transfer from YZ. Since much larger changes in the P680 + reduction kinetics of intact PS II occurred at comparable free-Ca2 + concentrations in the presence of EGTA, we conclude that EGTA influenced the P680 + reduction kinetics by directly interacting with PS II rather than by lowering the free Ca2 + concentration of the surrounding media. Notwithstanding these effects, we show that useful information about Ca2 + binding to PS II can be obtained when direct interaction of EGTA is taken into account. D 2003 Elsevier B.V. All rights reserved.

S

S

S

S

S

S

S

S

S

Keywords: Photosystem II; Ca2+ depletion; EGTA; Oxygen evolution; P680

1. Introduction In photosynthesis, the oxidation of water occurs at a manganese-containing water oxidizing center (WOC) of a multisubunit complex known as Photosystem II (PS II). The tetranuclear Mn cluster of the WOC accumulates oxidizing equivalents, with the redox states denoted by Sn for n={0, Abbreviations: AAS, atomic absorption spectroscopy; hDM, n-dodecyl h-D-maltoside; Chl, chlorophyll; DCBQ, 2,6-dichlorobenzoquinone; DIDS, diisothiocyano-stilbene-2,2V-disulfonic acid; EDTA, (ethylenedinitrilo)tetraacetic acid; EGTA, ethylene glycol bis (h-aminoethyl ether)-N,N,NV,NVtetraacetic acid; EPR, electron paramagnetic resonance; FWHM, full-width at half-maximum; MES, 2-(N-morpholino)ethanesulfonic acid; P680, primary electron donor of Photosystem II; PPBQ, phenyl-p-benzoquinone; PS II, Photosystem II; QA, primary quinone acceptor; RC, reaction center; TFP, trifluoperazine; YZ, tyr-161; WOC, water oxidizing center * Corresponding author. Tel.: +61-2-9413-7490; fax: +61-2-94137202. E-mail address: [email protected] (G.B. Stevens). 0005-2728/03/$ - see front matter D 2003 Elsevier B.V. All rights reserved. doi:10.1016/S0005-2728(03)00061-6

+

reduction

. . . , 4}. Changes in the S-state of the WOC are initiated by capture of a photon by chlorophyll (Chl) in a light-harvesting complex of PS II, the energy of which is transferred as an exciton to the primary electron donor of PS II, a chlorophyll dimer (P680) within the PS II reaction center (RC). After excitation, P680* decays exponentially with a time constant of 3 ps as it reduces a nearby pheophytin (I), the primary electron acceptor of PS II, forming the radical S S pair P680 +/I
. The radical pair decays with a time constant of 300 ps as it reduces a plastoquinone (QA), S forming the charge stable state P680 +/QA
. In O2-evolving S + PS II, P680 decays by oxidizing a tyrosine residue (YZ), S and in turn YZ+ oxidizes the Mn cluster at a rate that depends on the S-state of the WOC [1,2]. The role that calcium ion (Ca2 +) plays in the oxidation of water by PS II has been extensively studied over the recent years [3,4]. There is evidence that Ca2 + is required for the formation of the Mn cluster of the WOC [5– 8] and that it

22

G.B. Stevens, P.B. Lukins / Biochimica et Biophysica Acta 1605 (2003) 21–34

S

modulates electron transfer between YZ and P680 + [9– 13]. It has been shown to influence the oxidation of QA
[11,14] and the redox potential of QA [15 – 20], which was suggested to be a mechanism for modifying the charge recombination pathway [14,17 – 20] in order to protect PS II against photoinhibition [19] and during photoactivation of the WOC [18]. There is also evidence that Ca2 + influences the redox potential of cytochrome b-559 [21] and of the Mn cluster [22], that it ligates substrate H2O or OH
[23,24] and influences accessibility of substrate H2O to the WOC [25 – 27]. However, work still needs to be done to fully understand the involvement of Ca2 + in the molecular machinery of PS II [8,28]. Early evidence of the demand for Ca2 + in O2 evolution was obtained using PS II preparations made from spinach in which the 17 and 23 kDa extrinsic polypeptides had been removed by incubation with 2 M NaCl solution. Electron paramagnetic resonance (EPR) measurements showed that after this treatment, electron transfer from the Mn cluster to S YZ+ was inhibited. The NaCl treatment also caused a substantial loss of the rate of O2 evolution, which was S largely restored, together with YZ+ reduction from the Mn cluster, by Ca2 + at millimolar concentrations. Thus, Ca2 + was shown to be an essential cofactor in electron transfer S from the Mn cluster to YZ+ during O2 evolution of PS II [10,29]. The loss of O2-evolving activity was further shown to be due to loss of Ca 2 + rather than the 17 and 23 kDa polypeptides. Removal of bound Ca2 + from PS II preparations after incubation with NaCl by dialysis against the chelator ethylene glycol bis (h-aminoethyl ether)-N,N,NV,NVtetraacetic acid (EGTA), prevented the restoration of O2 evolution even after rebinding of the polypeptides had occurred [30]. However, the rate of O2 evolution of PS II with the 17 and 23 kDa extrinsic polypeptides intact remained largely unaffected in the presence of EGTA [30]. Thus, the 17 and 23 kDa extrinsic polypeptides seemed to form a barrier between the lumen and a Ca2 + binding site on PS II that influenced O2 evolution [3] or modulated the affinity of the Ca2 + binding site [31]. Transient absorption measurements on PS II preparations from higher plants showed that incubation with NaCl and S EGTA inhibited electron transfer from YZ to P680 + after formation of the S3 state [9]. Electron transfer from YZ to S P680 + was restored in the presence of Ca2 + at millimolar concentrations. Thus, in addition to influencing electron S transfer from the Mn cluster to YZ +, there is evidence that S 2+ Ca influences electron transfer from YZ to P680 + [9– 2+ 13]. More recently, it has been suggested that Ca influS ences electron transfer from YZ to P680 + by modifying the pK of a base cluster B near YZ. In this case, electron transfer S from YZ to P680 + would be inhibited, since in the absence 2+ of Ca , B would be inhibited from accepting a proton during the oxidation of water [12]. It was also proposed that Ca2 + may have a structural role in oxygen evolution by inducing an active conformation that

optimizes O2 evolution [32,33]. Changes in the rate of S P680 + decay in the presence of millimolar CaCl2 suggests that Ca2 + binding modifies the conformation of PS II near YZ [12,31]. In order to investigate the role of Ca2 + in S influencing electron transfer from YZ to P680 +, we meaS + sured P680 reduction kinetics of PS II membrane fragments in the presence of CaCl2 or EGTA, which were added to vary the Ca2 + concentration from 5 AM to 30 mM. Although EGTA does not affect the rate of O2 evolution of intact PS II [30], it is possible that EGTA itself may inhibit S electron transfer from YZ to P680 +. To eliminate possible interference from the effects of EGTA, we also measured the S P680 + reduction kinetics and O2 evolution of PS II core complexes in which Ca2 + was reduced by serial dilution in a Ca2 +-free buffer. Notwithstanding these mechanistic issues, we have been able to measure and interpret results on Ca2 + concentration S dependencies of P680 + reduction and O2 evolution to further clarify the role and mechanisms of Ca2 + function in PS II.

2. Materials and methods PS II-enriched thylakoid membrane fragments (BBYs) were isolated from spinach (Spinacia oleracea) according to the method of Berthold et al. [34] with modifications. After solubilizing in Triton X-100, PS II membrane fragments were washed once and resuspended in Ca2 +-free buffer (SMM). The SMM buffer was made up of 400 mM lowCa2 + sucrose (Sigma Ultra), 40 mM MES and 5 mM MgCl2 in Milli-Q water. The pH was adjusted to 6.5 by adding MgO. Extraneous MgO was removed by filtration through a 0.45 Am membrane. The concentration of contaminating Ca2 + in the buffer was determined by AAS to be less than 1 AM. To avoid Ca2 + contamination, plastic utensils were used which had been soaked for 24 h in 2% nitric acid followed by 24 h in 2% Extran MA03 solution (Merck) and finally rinsed with Milli-Q water. PS II core complexes were prepared from BBYs using the method of van Leeuwen et al. [35]. Surface-bound Ca2 + was removed from PS II core complexes by serial dilution. Concentration of the suspension of PS II core complexes was achieved by ultrafiltration under pressure using a stirred cell (Amicon model 8050) with a 300 kDa cutoff filter (YM300). The PS II core complexes were suspended at a concentration of 50 AM Chl in SMM buffer with 30 mM MgCl2 and 0.03% n-dodecyl h-D-maltoside (hDM) to prevent aggregation. Pressure was applied using N2 at 10 kPa above atmospheric pressure, while the suspension was kept at 4 jC in near-darkness. Several dilutions were required to reduce the free Ca2 + concentration to 15 AM. After the treatment, the sample was divided up into equal aliquots and S the steady-state O2 evolution and flash-induced P680 + reduction kinetics were measured for a range of free Ca2 + concentrations obtained by adding Ca2 + to these aliquots.

G.B. Stevens, P.B. Lukins / Biochimica et Biophysica Acta 1605 (2003) 21–34

The chlorophyll composition was calculated from absorption measurements of the preparations [36]. The concentration of free Ca2 + was measured using a Ca2 + ionselective electrode (Cole-Parmer 27502-08). Since the Ca2 + in the electrode membrane slowly equilibrates with that in solution, measurements were recorded after soaking the electrode in a solution at a similar concentration to the one being measured. The voltage obtained for a PS II suspension was converted to a free Ca2 + concentration by comparison with a calibration curve, which had been made S immediately before each P680 +/O2 measurement. The calibration curve was a non-linear least squares fit to electrode voltage as a function of Ca2 + concentration of solutions of known concentration. The rate of O2 evolution was measured using a Clark-type electrode (Hansatech DW1 or Lazar DO-166-MT) with saturating illumination from a 150 W quartz-halogen lamp, in the presence of 2 mM K3Fe(CN)6 and 0.32 mM PPBQ as electron acceptors. S P680 + reduction kinetics were measured using the 830 nm transient absorption apparatus described by Lukins et al. [37]. PS II suspensions were excited by 532 nm flashes with a full-width at half-maximum (FWHM) of 4 ns from a Qswitched Nd:YAG laser and the time dependence of the absorption at 830 nm monitored with 1 or 200 ns time resolution. The signal to noise ratio was improved by taking the average of 50 transient absorption measurements at a frequency of 2 Hz. Each PS II sample was typically exposed to 150 flashes before being discarded. Single-flash measurements of five flashes were taken of dark-adapted samples at a frequency of 20 Hz. The transient absorption decays were analyzed by fitting a sum of exponential decays using a Levenberg– Marquardt routine (Microcal Origin 6.0). The amplitudes and decay times of the fitted functions were not constrained during the fitting procedure.

3. Results 3.1. Adjustment of free Ca2+ in a suspension of PS II membranes with EGTA BBYs were suspended in SMM buffer and the free Ca2 + concentration was lowered by adding an aqueous solution of the acid form of EGTA and increased by adding CaCl2. The Cl
concentration was maintained at 60 mM by addition of MgCl2. The effect of EGTA on the free Ca2 + concentration of BBYs is shown in Fig. 1. After suspending the concentrated PS II preparation in Ca2 +-free buffer, the free Ca2 + concentration was 350 AM. Most of the free Ca2 + was introduced by the concentrated PS II preparation, since the Ca2 + contamination of SMM was less than 1 AM. Based on the chlorophyll concentration of the BBY sample (88 AM) and assuming 250 Chl per RC, an average of 994 Ca2 + ions were released by each PS II particle after resuspension of the concentrated PS II preparation in the SMM buffer. Thus, a large number of loosely bound Ca2 + ions were associated

23

Fig. 1. Effect of the added EGTA concentration on the free Ca2 + concentration of BBYs, suspended at a concentration of 88 AM Chl in SMM with 1 mM K3Fe(CN)6. Using K1 = 27,444 AM
1, the following parameters were obtained when fitting Eq. (2) to the data: TP = 374 F 39 AM, K2 = 0.057 F 0.016 AM
1 and Cmax = 282 F 12 AM.

with each PS II complex, even though the PS II had been isolated using a Ca2 +-free buffer. The relationship between the free Ca2 + concentration and the total EGTA concentration was analyzed using a model for Ca2 + binding to a single class of sites on PS II and a single class of Ca2 + binding sites on EGTA. In the model, the concentration of unoccupied binding sites on PS II for Ca2 + [ P], the free Ca2 + concentration [C] and the free EGTA concentration [E] were assumed to be at equilibrium: P þ CZPC and E þ CZEC

ð1Þ

The total concentration of Ca2 + (TC) and the total concentration of Ca2 + binding sites on PS II (TP) and EGTA (T E ) were defined by: T C =[C]+[PC]+[EC], TP=[ P]+[PC] and TE=[E]+[EC]. These equations, along with equilibrium binding constants K1 ¼ ð½PCÞ=ð½C½PÞ and K2 ¼ ð½ECÞ=ð½C½EÞ were solved to obtain the total concentration of binding sites on EGTA (TE) as a function of [C], K1, K2, TC and TP:    TP K1 ½C 1 TE ¼ TC
½C
1þ ; ð2Þ 1 þ K1 ½C K2 ½C with the boundary condition that at TE = 0, [C] = Cmax, so that TC=Cmax+(TPK1Cmax)/(1+K1Cmax). The value of TE was twice that of the concentration of added EGTA, since there are two Ca2 + binding sites on an EGTA molecule. In fitting the model to the data, K2 was fixed at the published value [38] and the other parameters were obtained from a non-linear least squares fit to the data. As seen from Fig. 1, Eq. (2) provides a reasonably good fit to the data. The binding affinity of the Ca2 + binding sites on PS II (1/K1), estimated from the best fit to the data in Fig. 1, was 19.1 F 5.4 AM and the concentration of Ca2 + binding

24

G.B. Stevens, P.B. Lukins / Biochimica et Biophysica Acta 1605 (2003) 21–34

sites on PS II (TP) was 374 F 39 AM. Based on 250 Chl per PS II complex, there was an average of 1064 F 110 Ca2 + binding sites on the surface of each PS II complex. 3.2. Effect of EGTA on P680 membranes

S+ reduction kinetics of PS II

Typical flash-induced 830 nm transient absorption measurements of the PS II membrane suspension is shown in Fig. 2. The increase in the transient absorption was due to the S formation of P680 + [39]. The maximum absorption change (0.0125) occurred at the peak of the flash intensity and was the same regardless of the presence of EGTA or Ca2 +. In these experiments, the PS II suspension was excited by

Fig. 2. Normalized flash-induced 830 nm transient absorption of BBY preparations, suspended at a Chl concentration of 88 AM with 2 mM K3Fe(CN)6 as an electron acceptor, with (A) nanosecond and (B) microsecond resolution. With no additives, the free Ca2 + concentration was 350 AM (middle traces). In the presence of 230 AM EGTA, the free Ca2 + concentration was 10 AM (top traces). After addition of CaCl2, the free Ca2 + concentration was 30 mM (bottom traces). Before making the 830 nm transient absorption measurements, BBY suspensions were passed through a 0.45 Am filter membrane to reduce scattering of the 830 nm probe beam by aggregated membranes.

flashes with an energy density of 0.4 mJ cm
2. At this S energy density, the amplitude of P680 + with exponential decay lifetimes of between 40 and 370 ns was 95% of that produced at an energy density of approximately 1.0 mJ cm
2. As there was no further increase in the extent of S P680 + formation at energy densities greater than 1.0 mJ
2 cm , the flashes were almost saturating. At 95% saturation, there was little residual energy in the flash to overcome S any inhibition in P680 + formation due to the presence of EGTA. We conclude that energy transfer from the antenna chlorophyll to the RC was unaffected by changes in the concentration of either Ca2 + or EGTA. The data in Fig. 2 shows that in the absence of added S Ca2 + or EGTA, the P680 + population decayed with approximately 70% nanosecond lifetimes and 30% microsecond lifetimes, which is typical for PS II membranes [32]. The addition of 30 mM Ca2 + resulted in a small increase in S the extent of the nanosecond P680 + decay compared to the control, in accordance with a previous report [32]. In the presence of 230 AM EGTA, the free Ca2 + concentration S decreased to 10 AM and P680 + decayed with about 30% nanosecond lifetimes and 70% microsecond lifetimes. In order to check whether these changes were reversible, the Ca2 + concentration was increased from 10 AM to 30 mM by S adding CaCl2 and the P680 + reduction kinetics were measured again. In this case, the extent of the nanosecond decay returned to that of the control, indicating that the kinetic changes were reversible and no significant photodamage had occurred to the PS II in the presence of EGTA. S The increase in the extent of microsecond P680 + decay amplitudes at the expense of nanosecond amplitudes is qualitatively similar to that observed with inside – out thylakoids after NaCl washing [40], with spinach PS II membranes after NaCl/EGTA washing [9] and with pea PS II core complexes after addition of (ethylenedinitrilo)tetraacetic acid (EDTA) [12]. We then measured the 830 nm transient absorption as a function of free-Ca2 + and EGTA concentration in order to observe the concentration trends. The decay of the 830 nm absorption change of BBY suspensions in the presence of varying amounts of EGTA and added Ca2 + was analyzed by fitting a sum of five exponential decays. Before fitting, the nanosecond traces were normalized and the absorption change in the microsecond region was scaled to the corresponding trace with nanosecond kinetics. The fitted parameters were obtained from unconstrained fits where no prior assumptions were made concerning the kinetic trends. S Free fits to the P680 + decay traces had approximately constant decay lifetimes but varying amplitudes, which are shown in Fig. 3. As the free Ca2 + concentration was S lowered, the amplitude of nanosecond P680 + decay kinetics, which usually indicate normal forward electron transfer to S P680 + [41], decreased while the amplitude of microsecond kinetics, which indicate charge recombination or slow forward electron transfer [42 –44], increased. The decay times varied slightly, but these variations were not statistically

G.B. Stevens, P.B. Lukins / Biochimica et Biophysica Acta 1605 (2003) 21–34

25

134 – 219 As, which are associated with slow or inhibited YZ S S to P680 + electron transfer and P680 +/QA
recombination, respectively. 3.3. Effect of EGTA on O2 evolution of PS II membranes In the presence of 500 AM EGTA, we observed that the rate of O2 evolution decreased by 20 F 13% to 422 F 36 Amol O2 (mg Chl)
1 h
1. A small decrease in the extent of O2 evolution was also observed when thylakoids that had been washed with a low-osmotic buffer were suspended in the presence of 500 mM EDTA [45] and when PS II membranes were suspended with A23187 and EGTA [30]. However, the O2 evolution of untreated thylakoids was not affected by EDTA [46]. S As shown in Fig. 3, the P680 + reduction amplitudes with lifetimes in the range 134 –219 As, which are assigned S to P680 +/QA
recombination in O2 inactive centers, increased by 30% in the presence of 272 AM EGTA. The extent of the increase in PS II centers with these decay lifetimes is comparable to the decrease in the extent of O2 evolution in the presence of EGTA.

S

3.4. P680 + reduction kinetics of PS II cores treated by serial dilution

S

Fig. 3. Relative amplitudes of P680 + decay components of BBYs suspended with EGTA or added Ca2 +: (A) 41 – 77 ns (D), 243 – 365 ns (+) and 6 As (5); and (B) 20 – 59 As (w ), 134 – 219 As (  ) and a DC offset representing components >1 ms (o).

significant. At Ca2 + concentrations above 350 AM, the S P680 + decay was dominated by exponential decay lifetimes of 41– 77 and 245 –365 ns. These times are associated with S P680 + reduction due to electron transfer from YZ in O2evolving PS II [41]. Variations in the decay lifetimes may be due to Coulombic attraction as the WOC becomes increasingly photooxidized in the four-step cycle of water oxidation [41]. Decay with lifetimes of 36 and 72 ns occur when the WOC is in the S0 and S1 state and with decay lifetimes of 58 and 406 ns when the WOC is in the S2 and S3 state [41]. Under the multiple flash conditions of this experiment, the PS II will be in a mixture of S-states, which accounts for the multiphasic decay of the absorption change. At Ca2 + conS centrations below f 100 AM, the P680 + decay was dominated by components with decay lifetimes of 20 –59 and

It was initially assumed that addition of EGTA would simply remove free Ca2 + and not interact directly with PS II. However, the data in Fig. 3 shows that significant S changes in the extent of nanosecond P680 + decay only occurred in the presence of EGTA. Consequently, part of the S change in the P680 + decay kinetics may have been due to direct interaction of EGTA with PS II centers. We tried to determine the extent to which free Ca2 + S affects the P680 + decay kinetics by reducing the free Ca2 + concentration by incubation with Chelex beads. However, Chelex caused the BBY preparation to aggregate, making it unsuitable for transient absorption measurements. Also, Chelex replaces Ca2 + with Na+, which has been shown to S increase the extent of the microsecond P680 + decay (Stevens, G.B. and Lukins, P.B. unpublished data). Therefore, serial dilution was used to remove free Ca2 + since this method does not require chelators, nor does it introduce ions that may interfere with the measurements. Unfortunately, treating BBYs by serial dilution caused the suspensions to aggregate, making them unsuitable for transient absorption measurements. This problem was overcome by using PS II core complexes which did not aggregate excessively. As shown in Fig. 4(A), the peak in the 830 nm transient absorption was the same at high and low free Ca2 + concentrations. Therefore, the absence of bound Ca2 + did not impede the flow of energy from the antenna to P680. S More importantly, the P680 + reduction kinetics of PS II cores in the presence of 30 AM free Ca2 + were similar to the S P680 + decay after addition of 30 mM Ca2 + and of untreated PS II cores [37].

26

G.B. Stevens, P.B. Lukins / Biochimica et Biophysica Acta 1605 (2003) 21–34

S

3.5. Comparison of P680 + reduction and O2 evolution in PS II core complexes after serial dilution The rate of O2 evolution of untreated PS II core complexes was 1260 Amol O2 (mg Chl)
1 h
1. The optical properties of serial diluted PS II samples are not ideal and cause substantially more noise in the 830 nm absorption transients compared to untreated samples. The data in Fig. 4(B) clearly shows that the decrease in the extent of O2 evolution was much greater than the decrease in the extent S of the nanosecond P680 + reduction with nanosecond lifetimes. This implies that these quantities are influenced by different types of PS II binding sites. More importantly, the loss of the ability of PS II to evolve O2 after serial dilution treatment was not accompanied by a significant change in S the P680 + reduction kinetics since these kinetics are similar to those of untreated PS II cores [37]. A loss of O2 evolution without a comparable change in the extent of nanosecond S P680 + decay was previously observed in PS II isolated from a thermophilic cyanobacteria that was purified by anion exchange chromatography [33] and in spinach PS II suspended with diisothiocyanostilbene-2,2V-disulfonic acid (DIDS) and trifluoperazine (TFP) [47]. In the second case, it was suggested that this effect was due to water oxidation to a peroxide intermediate that was released prematurely due to the destabilizing influence of DIDS or TFP [47]. 3.6. Binding constants for Ca2+ binding sites on PS II

S

Fig. 4. (A) Flash-induced transient 830 nm absorption of PS II cores, treated by serial dilution, suspended at a Chl concentration of 50 AM in SMM buffer with 5 mM K3Fe(CN)6 and free Ca2 + concentrations of 30 AM (top trace) and 30 mM (bottom trace). (B) Extent of P680 + decay with decay times of 10 – 107 ns (E) and extent of O2 evolution rates (.) of PS II core complexes treated by serial dilution, with a fit to Eq. (3), assuming Ca2 + binding at two classes of non-equivalent sites. The corresponding rates of O2 evolution were normalized to the rate at 3 mM Ca2 +.

S

The parameters that gave a satisfactory fit to the nanoS second P680 + reduction kinetics are shown in Fig. 4(B). The exponential lifetimes of the decays were in the range 10 –107 ns, which are typical for O2-evolving PS II cores [37]. Fig. 4(B) shows that the total amplitudes of the nanosecond kinetics reduces by f 20% as the Ca2 + concentration was reduced from 30 mM to 15 AM. The difference in the extent of the nanosecond kinetics was much less than the extent of the change in nanosecond kinetics of BBYs (60%) in the presence of EGTA (Fig. 3). This suggests that most of the change in the decay kinetics of BBYs treated with EGTA was due to the interaction of EGTA with PS II or that the binding affinity for a Ca2 + site S on PS II that influences nanosecond P680 + decay is higher in PS II cores than in BBYs.

A Hill plot of the extent of nanosecond P680 + decay is shown in Fig. 5(A). The data was fitted with a function that describes the extent of change in a variable (h) that indicates the fraction of PS II in a given state [48]: h¼

n1 K1 ½C n2 K2 ½C þ ; ðn1 þ n2 Þð1 þ K1 ½CÞ ðn1 þ n2 Þð1 þ K2 ½CÞ

ð3Þ

where Ki ¼ ð½Pi CÞ=ð½Pi ½CÞ is the binding constant and ni is the relative number of PS II binding sites of class Pi. The fit S to Eq. (3), in which h is the extent of nanosecond P680 + decay, is shown in Figs. 4(B) and 5(A). Assuming that the S extent of nanosecond P680 + decay was influenced only by 2+ Ca binding to PS II, the parameters of the fitted curve indicate that 90% of the change in the decay kinetics is influenced by a class of binding site with a dissociation constant (1/Ki) of 1.1 AM. This value is similar to that found for intact PS II membrane fragments [49]. Another class of binding site, with a dissociation constant of 3.7 mM, was also required for a good fit to the data. As expected for a Hill plot, the data falls within the regions bounded by the two lines with slope = 1 [48]. The upper line corresponds to the strong binding limit and the lower line to the weak-binding limit. As the concentration of free Ca2 + increases, the data points are closer to the region corresponding to weaker S binding. Assuming that P680 + kinetics are affected by 2+ Ca binding to PS II, this indicates either non-cooperative

G.B. Stevens, P.B. Lukins / Biochimica et Biophysica Acta 1605 (2003) 21–34

27

This dissociation constant compares favorably with values for high-affinity Ca2 + binding sites on PS II obtained with a range of assay methods for Ca2 + binding sites that influence PS II O2 evolution. Table 1 shows a summary of the Ca2 + binding constants for O2 evolution of a range of PS II preparations [31,49 – 51]. The values for the Ca2 + binding affinity vary because of differences in the methods used to obtain them, and possibly also because of differences in the PS II preparations themselves. Most of the entries in Table 1 represent two binding sites on the PS II, although there is also evidence that only one Ca2 + binding site affects O2 evolution [31,52]. 3.7. Effect of polypeptide composition on PS II Ca2+ binding sites

S

When the free Ca2 + concentration was decreased from S 30 mM to 15 –20 AM, the extent of P680 + decay with nanosecond lifetimes decreased by 58% in the case of BBYs in the presence of EGTA, and by 14% in the case of PS II cores treated by serial dilution. Assuming that the changes were due to Ca2 + binding, the difference in the extent of S nanosecond P680 + decay may be due to a higher Ca2 + binding affinity in PS II cores, arising from differences in the polypeptide composition [31]. To check this possibility, we measured the flash-induced 830 nm transient absorption of PS II cores in the presence of EGTA and analyzed S the P680 + reduction kinetics to determine the extent of S sub-microsecond P680 + decay for a range of free Ca2 + concentrations. As shown in Fig. 6, when the free Ca2 + concentration was lowered to 5 AM by the addition of 1 mM EGTA, the

Fig. 5. (A) Hill plot of the extent of 10 – 107 ns P680 + decay of PS II cores treated by serial dilution and fitted using Eq. (3) indicates binding sites with dissociation constants of 1.1 AM (n1 = 10) and 3.7 mM (n2 = 1). (B) Hill plot of the extent of O2 evolution fitted with Eq. (3) indicates binding sites with dissociation constants of 14 AM (n1 = 1) and 200 AM (n2 = 1). Also shown are the strong (upper dashed line) and weak (lower dashed line) binding limits.

Table 1 Dissociation constants of Ca2 + binding sites on PS II preparations that influence the restoration of O2 evolution Ca extraction

Dissociation constants (Ca/PS II)

Ca2 + determination

binding to two classes of Ca2 + binding sites or negative cooperative binding in which binding to the remaining sites becomes weaker at higher free Ca2 + concentrations. The Hill plot of the extent of O2 evolution, shown in Fig. 5(B), also suggests two sites with different binding affinities, although it could be argued that the data is scattered about a line with a slope of unity, which would indicate a single binding site. The extent of O2 evolution in Figs. 4(B) and 5(B) have also been fitted with Eq. (3). In this case, h is the extent of O2 evolution. A satisfactory fit to the data indicates that an equal number of binding sites, with dissociation constants of 14 and 200 AM influence the restoration of O2 evolution of the PS II cores. Alternatively, a reasonable fit to the data in Figs. 4(B) and 5(B) for the case of a single site has a dissociation constant of 55 AM.

PS II preparations, pH, extrinsic polypeptides [reference] Spinach core complex, pH 6.5 [this work] Wheat membranes
(17, 23) kDa, pH 6.0 [49] Spinach core complex, pH 6.0 [50] Spinach membranes
(17, 23) kDa, pH 6.5 [51] Spinach membranes +(17, 23) kDa, pH 6.5 [31] Spinach membranes
(17, 23) kDa, pH 6.75 [31]

serial dilution

55 AM or 14 AM and 200 AM 65 AM, 1.5 mM

Ca2 + electrode

2 M NaCl, EGTA/A23187 2M NaCl, EGTA/A23187

AAS

AAS

2M NaCl, EGTA/A23187

1 – 4 AM (1), 67 – 97 AM (1), 2.7 – 7 mM (1) 0.15 AM (2.5), 1.8 mM (4)

low pH/citrate NaCl/EGTA

60 AM (0.7), 1.7 mM (0.3)

2M NaCl, EGTA

26 AM (0.7), 0.5 mM (0.3)

scintillation counting 45 Ca2 + scintillation counting 45 Ca2 +

Ca2 + electrode

28

G.B. Stevens, P.B. Lukins / Biochimica et Biophysica Acta 1605 (2003) 21–34

compared to 5– 10 min incubation with EGTA, would be due to slow exchange of Ca2 + at the Ca2 + binding site on PS II as previously observed in PS II membranes [31]. 3.8. Effect of EGTA on nanosecond P680 S-state

S

Fig. 6. The extent of P680 + reduction kinetics with 12 – 109 ns lifetimes (E) and the extent of O2 evolution (.) of PS II core complexes suspended at a Chl concentration of 70 AM with 2 mM K3Fe(CN)6, 0.2 mM DCBQ and free Ca2 + concentration adjusted by the addition of EGTA or CaCl2.

S

extent of nanosecond P680 + decay decreased by 77%. As shown in Fig. 4(B), at comparable free Ca2 + concentrations, obtained by serial dilution, the extent of nanosecS ond P680 + decay decreased by 20%. Thus, the changes in the extent of nanosecond decay was strongly influenced by the presence of EGTA, presumably by solute interaction, and only slightly influenced by changes in the free Ca2 + concentration. The data in Fig. 6 also shows that the O2 evolution of spinach PS II cores was inhibited by 50 – 60% in the presence of 1 mM EGTA. The large decrease in the extent of O2 evolution is similar to the effect of EDTA on the O2 evolution of PS II preparations from cyanobacteria [46], and is larger than that in BBYs, probably because of the absence of the 17 and 23 kDa polypeptides. There is a good correlation between the flash-induced O2 evolution and S the extent of the nanosecond P680 + reduction kinetics of PS II cores in the presence of EGTA. Comparison of the data in Fig. 6 with that in Fig. 4(B) shows that, at the same free Ca2 + concentration, the extent of O2 evolution of PS II cores was lower after treatment by serial dilution than in the presence of EGTA. Of two possible explanations, it seems more likely that loss of Ca2 + from PS II, rather than binding of EGTA onto the Mn cluster of PS II, caused the reduction in O2 evolution. This is because if EGTA inhibited the O2 evolution by binding onto the Mn cluster, a greater reduction in O2 evolution would be expected in the PS II treated by EGTA as compared with serial dilution. As this was not the case, the loss of O2 evolution occurred as a result of the low free Ca2 + concentration in the suspension, due to addition of EGTA. We suggest that the lower rate and extent of O2 evolution after serial dilution, which took several hours

S+ decay for each

We carried out single-flash measurements to study possible inhibition of photooxidation of YZ and to see if there were other S-state-dependent effects. S The P680 + decay of PS II core complexes in the presence of 5 mM CaCl2 and 1 mM EGTA were measured for five flashes. As shown in Fig. 7(A), in the presence of 5 mM S CaCl2 there was a marked difference between the P680 + decay kinetics of the first and fifth flashes and those of the second, third and fourth flashes. This effect has been previously observed in PS II membranes [53] and PS II core S complexes [37]. The apparent slowing down of the P680 + reduction by YZ is thought to be due to the accumulation of a positive charge on the WOC [41]. A similar trend was also seen in the decay kinetics of PS II cores in the presence of EGTA, as shown in Fig. 7(B). S The P680 + decay kinetics were analyzed by fitting a sum of three exponential decays to both sets of traces. The amplitudes obtained by fitting the decays in Fig. 7 with fixed time constants, obtained after a survey of free fits, are shown in Fig. 8. The flash pattern of PS II cores in the presence of EGTA is clearly very similar to that of untreated

Fig. 7. Single-flash transient 830 nm absorption of dark-adapted PS II cores suspended at a Chl concentration of 100 AM with 2.5 mM K3Fe(CN)6 as an electron acceptor. The transients are the responses to the first flash (top traces) to the fifth flash (bottom traces) in the presence of either (A) 5 mM CaCl2 or (B) 1 mM EGTA. The flash repetition rate was 20 Hz and the curves are for single laser pulses.

G.B. Stevens, P.B. Lukins / Biochimica et Biophysica Acta 1605 (2003) 21–34

29

4. Discussion

Fig. 8. Amplitudes of single-flash P680S + reduction kinetics of PS II core complexes suspended with 5 mM CaCl2 (hollow symbols) or 1 mM EGTA (solid symbols): (A) DC offset representing components >1 As (w, x) and 25 ns (o, ); (B) 60 ns (5, n) and 170 ns (D, E).

.

PS II cores, although the decay with microsecond or longer decay times is about 50% higher than without EGTA. This data, together with a comparison of the amplitude of the microsecond component (50%) in the presence of EGTA after five flashes with the microsecond component of PS II cores after multiple flashes in Fig. 6 (approximately 80%), suggests that in the presence of EGTA, the microsecond amplitude increases gradually with each flash, possibly due to photoinhibition. The amplitude trend with flash number for time constants of 60 and 170 ns, shown in Fig. 8(B), was very similar to those of 40 and 95 ns previously reported for PS II cores [37]. The persistence of the S-state dependence of the kinetics agrees with the observed residual O2-evolving capacity of PS II cores in the presence of EGTA, as shown in Fig. 6. These experiments show that EGTA doesn’t affect any particular S-state of PS II but rather causes a reduction in the total nanosecond kinetic amplitudes.

We investigated the effect of added EGTA on the free Ca2 + concentration of suspensions of O2-evolving PS II membranes (BBYs). As far as we are aware, this is the first time that the binding equilibrium constant and number of non-specific surface binding sites per PS II complex has been estimated. The relationship between free Ca2 + and added EGTA to a PS II preparation was analyzed using a model of a single class of binding sites to PS II. The binding affinity and the number of surface binding sites were both higher than expected, since O2-evolving spinach PS II can function with two to three low affinity Ca2 + binding sites with a Km of 1.5 mM [49]. However, the number of binding sites on each PS II complex, made using a Ca2 +-free buffer, was in agreement with the number of Ca2 + released by the PS II preparation after it was suspended in Ca2 +-free buffer. Our primary goal was to investigate the effect of Ca2 + on S the P680 + reduction kinetics and O2 evolution of PS II membrane fragments. The flash-induced transient absorption at 830 nm increased rapidly due to the rapid formation S of P680 + and then decayed exponentially with lifetimes that depended on the free Ca2 + concentration of the PS II S suspension. The normalized amplitude of P680 + reduction kinetics with lifetimes of 41– 365 ns decreased from 70% to 30% after addition of 200 AM EGTA. Previous studies have S shown that EGTA and EDTA affect P680 + reduction kinetics in PS II after the 17 and 23 kDa extrinsic polypeptides have been removed [9,12]. However, this is the S first time that EGTA has been shown to affect P680 + decay in PS II with the 17 and 23 kDa polypeptides intact. As the EGTA concentration was increased, the free Ca2 + concenS tration decreased and nanosecond P680 + reduction was S + replaced by P680 reduction with lifetimes of 20 – 219 As. The data in Fig. 3 shows that at millimolar Ca2 + S concentrations, the ratio of amplitudes of P680 + exponential decay lifetimes A(41 – 47 ns):A(245 – 365 ns) was about 0.55, which is smaller than the ratio of comparable lifetimes obtained from repetitive flash measurements of spinach thylakoids, which had a ratio of approximately 1.48 [40]. Although there is no unique solution to fitting a noisy data set with a sum of exponential decays, we do expect better agreement between the ratios of the two fitted data sets. In S Ref. [40], the P680 + decay appears to have a significant amount of exponential decay with lifetimes of less than 10 S ns, which was not considered to reflect P680 + decay. The origin of this fast component was unclear but it was suggested to be due to the decay of 1Chl* or 3Chl* in the antenna or to decay of the radical pair. The existence of this signal in Ref. [40] indicates that the thylakoid membranes were overexcited by the laser flashes. In fitting the data with a sum of exponential decays, the rapidly decaying data was excluded. However, the ratio of fast to slow S nanosecond P680 + decay amplitudes is likely to have been higher than it would have been if the sample had not been overexcited in the first place. The data in Fig. 3 was

30

G.B. Stevens, P.B. Lukins / Biochimica et Biophysica Acta 1605 (2003) 21–34

S

obtained from P680 + decay data of PS II samples that had not been saturated by the laser flash and so the ratio of fast S to slow nanosecond P680 + decay amplitudes is probably closer to the true value. In a more recent analysis of the S P680 + decay of PS II core complexes excited by single flashes [37], an estimate of the ratio of fast to slow S nanosecond P680 + decay amplitudes was < 1, which is closer to the value obtained from the data in Fig. 3 than that in Ref. [40]. S The P680 + reduction kinetics with 134– 219 As lifetimes are assigned to charge recombination from QA
, which occurs in PS II with a disfunctional WOC [42 – 44]. The S assignment of P680 + reduction kinetics with lifetimes of approximately 35 As was originally made to inactive PS II centers [54]. More recently, flash-induced Chl fluorescence in combination with 830 nm transient absorption measurements of O2-evolving PS II demonstrated that the 35 As S P680 + lifetime oscillated with a periodicity of four and was therefore associated with O2 evolution [55]. Thus, at least S some of the P680 + decay with 20– 59 As lifetimes may be associated with O2-evolving centers. In this case, the loss of O2 evolution could be accounted for partly by the increase in the extent of the 134– 219 As decay kinetics. S The origin of the 35 As P680 + decay remains unclear, S although changes in the amplitude of P680 + decay kinetics when exchangeable protons are replaced by deuterons suggests that this component may be due to slow electron S transfer from YZ to P680 + [56,57]. Proposed models for this mechanism include a rate-limiting deprotonation of YZ S as a prerequisite for electron transfer to P680 + [12,56,57]. If this is the case, the observed increase in the amplitude of S the 35 As P680 + decay would suggest that deprotonation of YZ was inhibited by either loss of Ca2 + or the presence of EGTA. There is also recent evidence that the small 6 As component, which we found was not affected by changes in the free Ca2 + concentration, is also associated with O2 evolution [58]. The DC offset, which is assumed to represent S P680 + decay with lifetimes of >500 As, increased slightly in amplitude at low free Ca2 + concentrations. This component may be due to decay of P700+ from PS I complexes [40] which were present in small amounts in the BBY preparation. However, it could also be partly due to decay of triplet chlorophyll, formed by intersystem crossing from singlet excited chlorophyll in the antenna or RC of PS II, as intersystem crossing is also known to occur in isolated protein – pigment complexes [59]. S In contrast to the effect on the P680 + decay kinetics, EGTA did not significantly reduce the rate of O2 evolution of BBYs. This is in agreement with previous measurements of the effects of EGTA on the O2 evolution of PS II membranes [30]. A small decrease in the rate of O2 evolution was accompanied by a relatively large increase S in the amplitude of the 134 –219 As P680 + decay, which S + indicates P680 reduction by charge recombination beS S tween P680 + and QA
. Since P680 +/QA
charge recombi-

nation does not increase the S-state of the Mn center, PS II with these reaction kinetics would be incapable of evolving O2. However, it is possible for a significant amount of S P680 + decay by charge recombination from QA
without much loss in the rate of steady-state O2 evolution. This is S because the binding sites that influence the rate of P680 + decay will be in dynamic equilibrium with their surroundS ings. A PS II complex in which P680 + decays by charge
recombination from QA at one moment may have nanosecS ond P680 + decay the next because of changes at a binding S site that influences the rate of P680 + decay. Although the S extent of nanosecond P680 + decay after flash illumination may be slowed in a large number of RCs, under continuous illumination, there will be enough time during the oxidation of QA
, the rate-limiting step of O2 evolution in BBYs [60], for changes at the binding site to result in S-state turnover before QA
has been oxidized, and hence only a small effect on the rate of O2 evolution will be observed. S EGTA had a similar effect on the P680 + decay of PS II core complexes and BBYs, reducing the extent of nanosecS ond P680 + decay of PS II cores by 70%. However, EGTA also caused a significant decrease in the extent of O2 evolution (50%) of PS II cores. This indicates that the stability of the WOC is reduced in the absence of the 17 and 23 kDa proteins or that the reactivity of the WOC or its immediate environment with EGTA is enhanced in the case of PS II cores. Single-flash measurements on PS II core complexes showed that EGTA did not affect a particular Sstate of PS II but rather caused a uniform decrease in the extent of each component of nanosecond decay kinetics. This suggests that EGTA effects on the PS II donor side lead to a general inactivation of the WOC. In contrast to the effect of EGTA, treatment by serial dilution caused a small decrease (up to 20%) in the extent of nanosecond kinetics but a large decrease in the extent of O2 evolution, presumably because of a decrease in the stability of the PS II cores. The use of EGTA to remove Ca2 + seems to have caused an additional effect that was not evident when the more gentle method of serial dilution was used to reduce the Ca2 + concentration. Thus, the possibility that treatments used to induce a Ca2 + requirement in PS II had themselves caused an additional effect should be taken into account when interpreting the results of these experiments. This is a general concern which was previously emphasized in the study of the effects of chloride ion on photosynthetic oxygen evolution [61,62]. S The lack of a correlation between the P680 + reduction kinetics and the extent of O2 evolution suggests that PS II S binding sites that influence P680 + reduction kinetics are different from binding sites that influence O2 evolution. In S the absence of EGTA, the extent of nanosecond P680 + 2+ decay decreased by no more than 20% when the free Ca concentration was reduced by serial dilution to 15 AM (Fig. S 4(B)). In contrast, the extent of nanosecond P680 + decay 2+ decreased by 60% when the Ca concentration was reduced to the same level in the presence of EGTA (Fig.

G.B. Stevens, P.B. Lukins / Biochimica et Biophysica Acta 1605 (2003) 21–34

3(A)). This can be explained by either direct interaction of EGTA with PS II or a stronger Ca2 + binding affinity of the PS II cores because of the absence of the extrinsic polypeptides. The second suggestion is supported by evidence that Km values for Ca2 + binding to PS II which lack the 17 and 23 kDa polypeptides are about half those with the polypeptides intact [31]. However, a comparison of the data in Figs. 3 and 6 S shows that at hypothetical binding sites that regulate P680 + reduction, PS II cores and BBYs appear to have the same Ca2 + binding affinity. Thus, since Ca2 + binding does not appear to be stronger in PS II cores than in PS II membranes, we conclude that direct interaction of EGTA with PS II, rather than Ca2 + loss from PS II, was the principal cause S of the decrease in the extent of nanosecond P680 + decay. We suggest that basic protein residues lysine and arginine, which are positively charged at neutral pH, are possible binding sites for EGTA. At low free Ca2 + concentration, EGTA binds onto PS II resulting in a conformational S change that inhibits electron transfer from YZ to P680 +. Restoration of a conformation for rapid electron transfer S from YZ to P680 + as the free Ca2 + concentration was increased may be due to a combination of additional Ca2 + binding to PS II and EGTA de-binding from PS II and preferentially binding to free Ca2 + instead. The conclusion that EGTA causes a decrease in the S extent of nanosecond P680 + by directly binding onto PS II challenges that of a previous report in which a decrease in these kinetic components, observed after NaCl/EGTA-treatment of PS II membranes, was attributed to the absence of the 17 and 23 kDa polypeptides [9]. However, there is evidence that PS II preparations with and without the 17 and 23 kDa polypeptides do have the same extent of subS microsecond P680 + decay kinetics under similar conditions [37]. Thus, the loss of the 17 and 23 kDa polypeptides does not necessarily cause a decrease in the extent of nanosecond S S P680 + decay. The loss of the nanosecond P680 + decay kinetics reported earlier may instead have been due to direct binding of EGTA to PS II, which was used at a high concentration (20 mM) during the preparation of the reconstituted membranes [63]. Since direct binding of EGTA with PS II caused a S decrease in the nanosecond P680 + reduction kinetics of both BBYs and PS II cores, EGTA binding sites that S influence the extent of nanosecond P680 + decay would presumably be available for EGTA binding regardless of whether the 17 and 23 kDa polypeptides were present. Thus, the EGTA binding site would not be located under or in the immediate vicinity of the 17 and 23 kDa polypeptides. If EGTA binds near or onto the Mn center of 17 and 23 kDa depleted PS II as suggested by EPR, ESEEM and FTIR studies [63 – 67], this is likely to inhibit O2 evolution. In this case, the extent of the change in amplitude of flash-induced S P680 + decay indicating charge recombination via QA
would be expected to change by the same amount as the change in the rate of O2 evolution. However, since the

31

extent of the change in the rate of O2 evolution of BBYs S was much less than the change in the P680 + decay kinetics, it appears that an EGTA binding site on intact PS II that S influences the P680 + decay kinetics is different from a binding site that influences O2 evolution. S In order for P680 + to decay by electron transfer from YZ, S there should be sufficient time between flashes for YZ + to be reduced by the Mn complex. As shown in Fig. 4(B), the large S extent of nanosecond P680 + reduction kinetics indicates that YZ was mostly in a reduced state before each flash, regardless of the Ca2 + concentration. In the absence of S reduced YZ, P680 + is expected to decay with a time constant of approximately 200 As, due to charge recombination with QA
. This occurs when the Mn cluster is removed, resulting in the loss of O2 evolution. During normal O2 S evolution, YZ + decays exponentially with time constants of up to 1.5 ms, depending on the S-state of the Mn cluster. Thus, it is expected that under our experimental conditions of 500 ms between flashes, there would have been sufficient S time for the complete reduction of YZ + before each flash. However, as shown in Fig. 4(B), loss of O2 evolution occurred without a corresponding decrease in the extent of S nanosecond P680 + decay kinetics, which indicates that SS state turnover was inhibited but YZ + reduction was not. There are several possible explanations for this interestS ing effect. One is that in the absence of Ca2 +, YZ + may be reduced at a slower rate than normal, but rapidly enough to S enable YZ + reduction before the next flash. The observed reduction in the rate of O2 evolution would then be S explained by an increase in the decay rate of YZ + reduction
to greater than the decay times for QA oxidation, which occurs with times of 0.95 and 14.9 ms with approximately equal amplitude for intact BBY preparations [68]. However, in Ca2 +-depleted PS II membranes, obtained by washing S with NaCl/EGTA, YZ + decayed with a half-life of 2 s [69]. Therefore, it seems unlikely that the same kind of impairment had occurred to the PS II cores treated by serial S dilution, since YZ + decay with a half-life of 2 s would, in S our experiment, have resulted in P680 + decay by charge
recombination with QA . S It was possible that YZ + may have been reduced by ferrocyanide, which is assumed to be present at low concentration as a byproduct of the reduction of ferricyanide. S This was suggested as a possible reduction pathway for YZ + in Mn-depleted (Tris-treated) PS II in the presence of DCMU, which was added to block QA
oxidation [70]. S However, YZ + reduction with a decay time of less than 2 s by this pathway would be unlikely for PS II with the Mn S cluster intact because YZ + of a salt-washed BBY preparation in the presence of 1 mM Fe(CN)64
decayed with a time constant of 2 s [69]. As suggested previously [47], premature release of a peroxide intermediate of water oxidation which evidently forms before turnover to S4 [71] may account for the low rate of O2 evolution in PS II with unimpeded electron S transfer to P680 +. This assignment is supported by obser-

32

G.B. Stevens, P.B. Lukins / Biochimica et Biophysica Acta 1605 (2003) 21–34

vations of the formation and release of peroxide accompanied by a loss of O2 evolution in PS II preparations after washing with molar NaCl or CaCl2 or after subjecting PS II to osmotic stress or lipophilic compounds [72]. Since these effects are partially reversed after addition of CaCl2, this interpretation of our results would support a hypothesis for Ca2 + as a binding site for substrate H2O or OH
[23,24] or as a regulator for the accessibility of substrate water to the WOC [25 – 27]. Our analysis suggests that the decrease in O2-evolving ability was due to loss of Ca2 +, either from two sites with dissociation constants of 14 and 200 AM or from a single site with a dissociation constant of 55 AM. The absence of a S change in the P680 + reduction kinetics at low free Ca2 + concentrations implies that an exogenous electron source S that reduces YZ + between flashes was unaffected by the loss of Ca2 +. Since O2 evolution of BBYs is hardly affected at low free-Ca2 + concentrations, whereas the O2 evolution of PS II cores is affected at similar free-Ca2 + concentrations, regardless of the method used to lower the free Ca2 + concentration, the Ca2 +-binding site that influences the extent of O2 evolution may be shielded by the 17 and 23 kDa polypeptides. At this site, Ca2 + is probably bound by carboxylate residues of aspartate and glutamate in the lumenal interhelical domains and the carboxy terminal domain of the D1 polypeptide [73]. Finally, a number of previous reports suggest that Ca2+ is S required for rapid electron transfer from YZ to P680 + [9– 2+ 13]. After serial dilution treatment, the free Ca concentration had been reduced to a concentration of 15 AM. According to the hypothesis in which electron transfer from S YZ to P680 + depends on YZ being in a deprotonated state [12,55,56], our results suggest that EGTA rather than Ca2 + may have affected the ability of base cluster B [12] to accept a proton from YZ, since Ca2 + loss from the PS II suspended in a free Ca2 + concentration of 15 AM in the absence of EGTA did not result in a significant reduction in the extent S of the nanosecond P680 + decay.

Acknowledgements We thank Chris Connolly for the AAS measurements, and Tony Larkum, Tom Wydrzynski, Warwick Hillier, Ray Ritchie, Adele Post and Gert Christen for helpful discussions. Financial support from the Australian Research Council is gratefully acknowledged.

References [1] J. Barber, Photosystem II: a multisubunit membrane protein that oxidises water, Curr. Opin. Struct. Biol. 12 (2002) 523 – 530. ¨ . Hansson, T. Wydrzynski, Current perceptions of Photosystem II, [2] O Photosynth. Res. 23 (1990) 131 – 162.

[3] C.F. Yocum, Calcium activation of photosynthetic water oxidation, Biochim. Biophys. Acta 1059 (1991) 1 – 15. [4] R.J. Debus, The manganese and calcium ions of photosynthetic oxygen evolution, Biochim. Biophys. Acta 1102 (1992) 269 – 352. [5] L. Zaltsman, G.M. Ananyev, E. Bruntrager, G.C. Dismukes, Quantitative kinetic model for photoassembly of the photosynthetic water oxidase from its inorganic constituents: Requirements for manganese and calcium in the kinetically resolved steps, Biochemistry 36 (1997) 8914 – 8922. [6] G.M. Ananyev, G.C. Dismukes, Calcium induces binding and formation of a spin-coupled dimanganese(II,II) center in the apo-water oxidation complex of Photosystem II as precursor to the functional tetra-Mn/Ca cluster, Biochemistry 36 (1997) 11342 – 11350. [7] R.M. Cinco, K.L. McFarlane Holman, J.H. Robblee, J. Yano, S.A. Pizarro, E. Bellacchio, K. Sauer, V.K. Yachandra, Ca2 + EXAFS establishes the Mn – Ca cluster in the oxygen-evolving complex of Photosystem II, Biochemistry 41 (2002) 12928 – 12933. [8] G.M. Ananyev, L. Zaltsman, C. Vasko, G.C. Dismukes, The inorganic biochemistry of photosynthetic oxygen evolution/water oxidation, Biochim. Biophys. Acta 1503 (2001) 52 – 68. [9] A. Boussac, P. Se´tif, A.W. Rutherford, Inhibition of tyrosine Z photooxidation after formation of the S3 state in Ca2 +-depleted and Cl
depleted Photosystem II, Biochemistry 31 (1992) 1224 – 1234. [10] A. Bakou, D.F. Ghanotakis, Substitution of lanthanides at the calcium site(s) in Photosystem II affects electron transport from tyrosine Z to P680+, Biochim. Biophys. Acta 1141 (1992) 303 – 308. [11] L.-E. Andre´asson, I. Vass, S. Styring, Ca2 + depletion modifies the electron transfer on both donor and acceptor sides in Photosystem II from spinach, Biochim. Biophys. Acta 1230 (1995) 155 – 164. [12] M. Haumann, W. Junge, Evidence for impaired hydrogen-bonding of tyrosine YZ in calcium-depleted Photosystem II, Biochim. Biophys. Acta 1411 (1999) 121 – 133. [13] R. Ahlbrink, B.K. Semin, A.Y. Mulkidjanian, W. Junge, Photosystem II of peas: effects of added divalent cations of Mn, Fe, Mg, and Ca on + two kinetic components of P680 reduction in Mn-depleted core particles, Biochim. Biophys. Acta 1506 (2001) 117 – 126. [14] A. Krieger, A.W. Rutherford, C. Jegerscho¨ld, Thermoluminescence measurements on chloride-depleted and calcium-depleted Photosystem II, Biochim. Biophys. Acta 1364 (1998) 46 – 54. [15] A. Krieger, E. Weis, Energy dependent quenching of chlorophyll-afluorescence: the involvement of proton-calcium exchange at Photosystem 2, Photosynthetica 27 (1992) 89 – 98. [16] A. Krieger, E. Weis, The role of calcium in the pH-dependent control of Photosystem II, Photosynth. Res. 37 (1993) 117 – 130. [17] A. Krieger, E. Weiss, S. Dementer, Low-pH-induced Ca2 + ion release in the water-splitting system is accompanied by a shift in the midpoint redox potential of the primary quinine acceptor QA, Biochim. Biophys. Acta 1144 (1993) 411 – 418. [18] G.N. Johnson, A.W. Rutherford, A. Krieger, A change in the midpoint potential of the quinine QA in Photosystem II associated with photoactivation of oxygen evolution, Biochim. Biophys. Acta 1229 (1995) 202 – 207. [19] A. Krieger, A.W. Rutherford, Comparison of chloride-depleted and calcium-depleted PS II: the midpoint potential of QA and susceptibility to photodamage, Biochim. Biophys. Acta 1319 (1997) 91 – 98. [20] N. Keren, I. Ohad, A.W. Rutherford, F. Drepper, A. Krieger-Liszkay, Inhibition of Photosystem II activity by saturating single turnover flashes in calcium-depleted and active Photosystem II, Photosynth. Res. 63 (2000) 209 – 216. [21] V.P. McNamara, K. Gounaris, Granal Photosystem II complexes contain only the high redox potential form of cytochrome b-559 which is stabilized by the ligation of calcium, Biochim. Biophys. Acta 1231 (1995) 289 – 296. [22] P.J. Riggs-Gelasco, R. Mei, D.F. Ghanotakis, C.F. Yocum, J.E. PennerHahn, X-ray absorption spectroscopy of calcium-substituted derivatives of the oxygen-evolving complex of Photosystem II, J. Am. Chem. Soc. 118 (1996) 2400 – 2410.

G.B. Stevens, P.B. Lukins / Biochimica et Biophysica Acta 1605 (2003) 21–34 [23] J.S. Vrettos, D.A. Stone, G.W. Brudvig, Quantifying the ion selectivity of the Ca2 + site in Photosystem II: evidence for direct involvement of Ca2 + in O2 formation, Biochemistry 40 (2001) 7937 – 7945. [24] J.S. Vrettos, J. Limburg, G.W. Brudvig, Mechanism of photosynthetic water oxidation: combining biophysical studies of Photosystem II with inorganic model chemistry, Biochim. Biophys. Acta 1503 (2001) 229 – 245. [25] J. Tso, M. Sivaraja, G.C. Dismukes, Calcium limits substrate accessibility or reactivity at the manganese cluster in photosynthetic water oxidation, Biochemistry 30 (1991) 4734 – 4739. [26] T. Wydrzynski, W. Hillier, J. Messinger, On the functional significance of substrate accessibility in the photosynthetic water oxidation mechanism, Physiol. Plant. 96 (1996) 342 – 350. [27] K.A. Vander Meulen, A. Hobson, C.F. Yocum, Calcium depletion modifies the structure of the Photosystem II O2-evolving complex, Biochemistry 41 (2002) 958 – 966. [28] P.H. Homann, Chloride and calcium in Photosystem II: from effects to enigma, Photosynth. Res. 73 (2002) 169 – 175. [29] D.F. Ghanotakis, G.T. Babcock, C.F. Yocum, Calcium reconstitutes high rates of oxygen evolution in polypeptide depleted Photosystem II preparations, FEBS Lett. 167 (1984) 127 – 130. [30] D.F. Ghanotakis, J.M. Topper, G.T. Babcock, C.F. Yocum, Watersoluble 17 and 23 kDa polypeptides restore oxygen evolution activity by creating a high-affinity binding site for Ca2 + on the oxidizing side of Photosystem II, FEBS Lett. 170 (1984) 169 – 173. ¨ delroth, K. Lindberg, L.-E. Andre´asson, Studies of Ca2 + binding [31] P. A in spinach Photosystem II using 45Ca2 +, Biochemistry 34 (1995) 9021 – 9027. [32] M. Vo¨lker, H.-J. Eckert, G. Renger, Effects of trypsin and bivalent cations on P-680+-reduction, fluorescence induction and oxygen evolution in Photosystem II membrane fragments from spinach, Biochim. Biophys. Acta 890 (1987) 66 – 76. [33] S. Pauly, E. Schlodder, H.T. Witt, The influence of salts on charge separation (P680+/Qa
) and water oxidation of Photosystem II complexes from thermophilic cyanobacteria. Active and inactive conformational states of Photosystem II, Biochim. Biophys. Acta 1099 (1992) 203 – 210. [34] D. Berthold, G. Babcock, C. Yocum, A highly-resolved oxygenevolving Photosystem II preparation form spinach thylakoid membranes, FEBS Lett. 134 (1981) 231 – 234. [35] P.J. van Leeuwen, M.C. Nieveen, E.J. van de Meent, J.P. Dekker, H.J. Gorkom, Rapid and simple isolation of pure Photosystem II core and RC particles from spinach, Photosynth. Res. 28 (1991) 149 – 153. [36] R.J. Porra, W.A. Thompson, P.E. Kriedemann, Determinataion of accurate extinction coefficients and simultaneous equations for assaying chlorophylls a and b extracted with four different solvents: verification of the concentration of chlorophyll standards by atomic absorption spectroscopy, Biochim. Biophys. Acta 975 (1989) 384 – 394. [37] P.B. Lukins, A. Post, P.J. Walker, A.W.D. Larkum, P680+ reduction in oxygen-evolving Photosystem II core complexes, Photosynth. Res. 49 (1996) 209 – 221. [38] A.E. Martell, R.M. Smith, Critical Stability Constants, Amino Acids, vol. 1, Plenum, New York, 1974, p. 269. [39] J.A. van Best, P. Mathis, Kinetics of the reduction of the oxidized primary electron donor of Photosystem II in spinach chloroplasts and in chlorella cells in the microsecond and nanosecond time ranges following flash excitation, Biochim. Biophys. Acta 503 (1978) 178 – 188. ˚ ckerlund, K. Brettel, H.T. Witt, Reversible alteration of nano[40] H.-E. A second reduction of chlorophyll aII+ in inside-out thylakoids correlated to inhibition and reconstitution of oxygen-evolving activity, Biochim. Biophys. Acta 765 (1984) 7 – 11. [41] K. Brettel, E. Schlodder, H.T. Witt, Nanosecond reduction kinetics of photooxidized chlorophyll-aII (P-680) in single flashes as a probe for the electron pathway, H+-release and charge accumulation in the O2evolving complex, Biochim. Biophys. Acta 766 (1984) 403 – 415.

33

[42] J. Haveman, P. Mathis, Flash-induced absorption changes of the primary donor of Photosystem II at 820 nm in chloroplasts inhibited by low pH or tris-treatment, Biochim. Biophys. Acta 440 (1976) 346 – 355. [43] G. Renger, C.H. Wolff, The existence of a high photochemical turnover rate at the reaction centres of system II in tris-washed chloroplasts, Biochim. Biophys. Acta 423 (1976) 610 – 614. [44] E. Schlodder, K. Brettel, H.T. Witt, Relation between microsecond reduction kinetics of photooxidized chlorophyll aII (P-680) and photosynthetic water oxidation, Biochim. Biophys. Acta 808 (1985) 123 – 131. [45] T. Wydrzynski, G. Renger, On the interpretation of the NMR waterproton relaxivity of photosynthetic membrane samples: ramifications in the use of EDTA, Biochim. Biophys. Acta 851 (1986) 65 – 74. [46] K. Satoh, S. Katoh, Inhibition by ethylenediamine tetraacetate and restoration by Mn2 + and Ca2 + of oxygen-evolving activity in Photosystem II preparation from the thermophilic cyanobacterium, Synechococcus sp., Biochim. Biophys. Acta 806 (1985) 221 – 229. [47] H.-J. Eckert, T. Wydrzynski, G. Renger, The effects of diisothiocyanostilbene-2,2V-disulfonic acid (DIDS), trifluoperazine and lauroylcholinechloride on P-680+ reduction and oxygen evolution, Biochim. Biophys. Acta 932 (1988) 240 – 249. [48] K.E. van Holde, W.C. Johnson, P.S. Ho, Principles of Physical Biochemistry, Prentice Hall, New Jersey, 1998, pp. 598 – 619. [49] K.V. Cammarata, G.M. Cheniae, Studies on 17, 24 kD depleted Photosystem II membranes, Plant Physiol. 84 (1987) 587 – 595. [50] K. Kalosaka, W.F. Beck, G. Brudvig, G. Cheniae, Coupling of the PS2 reaction center to the O2-evolving center requires a very high affinity Ca2 + site, in: M. Baltscheffsky (Ed.), Current Research in Photosynthesis, vol. 1, Kluwer Academic Publishers, Netherlands, 1990, pp. 721 – 724. [51] G. Grove, G.W. Brudvig, Calcium binding studies of Photosystem II using a calcium-sensitive electrode, Biochemistry 37 (1998) 1532 – 1539. [52] K.-C. Han, S. Katoh, Different localization of two Ca2 + in spinach oxygen-evolving Photosystem II membranes. Evidence for involvement of only one Ca2 + in oxygen evolution, Plant Cell Physiol. 34 (1993) 585 – 593. [53] H.-J. Eckert, G. Renger, Temperature dependence of P680+ reduction in O2-evolving PS II membrane fragments at different redox states Si of the water oxidizing system, FEBS Lett. 236 (1988) 425 – 431. ¨ . Hansson, Flash photolysis studies of [54] C.W. Hoganson, P.A. Casey, O manganese-depleted Photosystem II: evidence for binding of Mn2 + and other transition metal ions, Biochim. Biophys. Acta 1057 (1991) 399 – 406. [55] G. Christen, F. Reifarth, G. Renger, On the origin of the ‘35 As kinetics’ of P680 + reduction in Photosystem II with an intact water oxidising complex, FEBS Lett. 429 (1998) 49 – 52. [56] G. Christen, G. Renger, The role of hydrogen bonds for the multiphasic P680+ reduction by Yz in Photosystem II with intact oxygen evolution capacity. Analysis of kinetic H/D isotope exchange effects, Biochemistry 38 (1999) 2068 – 2077. [57] R. de Wijn, T. Schrama, H.J. van Gorkom, Secondary stabilization reactions and proton-coupled electron transport in Photosystem II investigated by electroluminescence and fluorescence spectroscopy, Biochemistry 40 (2001) 5821 – 5834. [58] M.J. Schilstra, F. Rappaport, J.H.A. Nugent, C.J. Barnett, D.R. Klug, Proton/hydrogen transfer affects the S-state-dependent microsecond phases of P680+ reduction during water splitting, Biochemistry 37 (1998) 3974 – 3981. [59] S. Santabarbara, E. Bordignon, R.C. Jennings, D. Carbonera, Chlorophyll triplet states associated with Photosystem II of thylakoids, Biochemistry 41 (2002) 8184 – 8194. [60] G. Renger, B. Hanssum, H. Gleiter, H. Koike, Y. Inoue, Interaction of 1,4-benzoquinones with Photosystem II in thylakoids and Photosystem II membrane fragments from spinach, Biochim. Biophys. Acta 936 (1988) 435 – 446.

S

S

34

G.B. Stevens, P.B. Lukins / Biochimica et Biophysica Acta 1605 (2003) 21–34

[61] K. Olesen, L.-E. Andre´asson, The function of the chloride ion in photosynthetic oxygen evolution, Biochemistry 42 (2003) 2025 – 2035. [62] T. Wydrzynski, F. Baumgart, F. MacMillan, G. Renger, Is there a direct chloride requirement in the oxygen-evolving reactions of Photosystem II, Photosynth. Res. 25 (1990) 59 – 72. [63] A. Boussac, J.-L. Zimmermann, A.W. Rutherford, EPR signals from modified charge accumulation states of the oxygen-evolving enzyme in Ca2 +-deficient Photosystem II, Biochemistry 28 (1989) 8984 – 8989. [64] A. Boussac, J.-L. Zimmermann, A.W. Rutherford, Factors influencing the formation of modified S2 EPR signal and the S3 EPR signal in Ca2 +-depleted Photosystem II, FEBS Lett. 277 (1990) 69 – 74. [65] J.-L. Zimmermann, A. Boussac, A.W. Rutherford, The manganese center of oxygen-evolving and Ca2 +-depleted Photosystem II: a pulsed EPR spectroscopy study, Biochemistry 32 (1993) 4831 – 4841. [66] P. van Vliet, A. Boussac, A.W. Rutherford, Chloride-depletion effects in the calcium-deficient oxygen-evolving complex of Photosystem II, Biochemistry 33 (1994) 12998 – 13004. [67] Y. Kimura, T.-A. Ono, Chelator-induced disappearance of carboxylate stretching vibrational modes in S2/S1 FTIR spectrum in oxygenevolving complex of Photosystem II, Biochemistry 40 (2001) 14061 – 14068. [68] G. Renger, H.-J. Eckert, A. Bergmann, J. Bernarding, B. Liu, A.

[69]

[70]

[71]

[72]

[73]

Napiwotzki, F. Reifarth, H.J. Eichler, Fluorescence and spectroscopic studies of exciton trapping and electron transfer in Photosystem II of higher plants, Aust. J. Plant Physiol. 22 (1995) 167 – 181. N. Lydakis-Simantiris, P. Dorlet, D.F. Ghanotakis, G.T. Babcock, Kinetic and spectroscopic properties of the YZS radical in Ca2 +- and Cl
-depleted Photosystem II preparations, Biochemistry 37 (1998) 6427 – 6435. J.P. Dekker, H.J. van Gorkom, M. Brock, L. Ouwehand, Optical characterization of Photosystem II electron donors, Biochim. Biophys. Acta 764 (1984) 301 – 309. V. Klimov, G. Ananyev, O. Zastryzhnaya, T. Wydrzynski, G. Renger, Photoproduction of hydrogen peroxide in Photosystem II membrane fragments: a comparison of four signals, Photosynth. Res. 38 (1993) 409 – 416. W. Hillier, T. Wydrzynski, Increases in peroxide formation by the Photosystem II oxygen evolving reactions upon removal of the extrinsic 16, 22 and 33 kDa proteins are reversed by CaCl2 addition, Photosynth. Res. 38 (1993) 417 – 423. H.-A. Chu, A.P. Nguyen, R. Debus, Amino acid residues that influence the binding of manganese or calcium to Photosystem II: 1. The luminal interhelical domains of the D1 polypeptide, Biochemistry 34 (1995) 5839 – 5858.