EPR Studies of Photosystem II

!" # $%&!$! "!%" &!" '()(*()) ) +,!$$-$

..V 2 ISBN

C D E E

F C

! "# $ " % # "$ &'()" "*+,'-(. "/

GHJ

K LK MM FM O P

Till Sven och Freja

List of Papers

This thesis is based on the following published papers and manuscripts, which are referred to in the text by their Roman numerals. I Havelius, K. G. V., Su, J.-H., Feyziyev, Y., Mamedov, F. and Styring, S. (2006) The spectral resolution of the split EPR- signals induced by illumination at 5 K from the S1, S3 and S0 states of photosystem II, Biochemistry, 45:9279–9290. II Su, J.-H., Havelius, K. G. V., Feyziyev, Y., Mamedov, F., Ho, F. M. and Styring S. (2006) Split EPR signals from photosystem II are modified by methanol, reflecting S state dependent binding and alterations in the magnetic coupling in the CaMn4 cluster, Biochemistry, 45:7617–7627. III Havelius, K. G. V. and Styring, S. (2007) pH dependent competition between YZ and YD in photosystem II probed by illumination at 5 K, Biochemistry, 46:7865–7874. IV Su, J.-H., Havelius, K. G. V., Ho, F. M., Han, G., Mamedov, F. and Styring, S. (2007) Formation spectra of the EPR split signal from S0, S1 and S3 States in photosystem II induced by monochromatic light at 5 K, Biochemistry, 46:10703-10712. V Han, G., Ho, F. M., Havelius, K. G. V., Morvaridi, S. F., Ma- medov, F. and Styring, S. (2008) Direct quantification of the four individual S states in photosystem II using EPR spectroscopy, Biochim. Biophys. Acta, 1777:496-503. VI Havelius, K. G. V., Ho, F. M., Su, J.-H., Han, G., Mamedov, F. and Styring, S. (2009) The same origin of the split EPR signal induced by visible or near-infrared light at liquid helium temperature from the S3-state of photosystem II, Manuscript VII Sjöholm, J., Havelius, K. G. V., Mamedov, F. and Styring, S. (2009) The S0 state of water oxidizing complex in photosystem II: pH dependence of the EPR split signal, induction and mechanistic implications, Manuscript VIII Ho, F. M., Havelius, K. G. V., Sjöholm, J. and Styring, S. (2009) Investigation and characterisation of EPR split signals in the native S2 state of photosystem II, Early manuscript Reprints were made with permission from the respective publishers.

The following publication is not included in this thesis and deals with the work of the master degree thesis (Filsofie Magister) published under my maiden name Sigfridsson.

IX Sigfridsson, K. G. V., Bernát, G., Mamedov, F. and Styring, S. (2004) Molecular interference of Cd2+ with photosystem II, Biochim. Biophys. Acta, 1659:19–31.

Contents

Introduction...... 9 1. Why do we study Photosystem II?...... 9 1.1. We need energy...... 9 1.2. Be biomimetic ...... 10 2. Oxygenic Photosynthesis ...... 11 2.1. Light reactions...... 11 2.3. Dark reactions (not light-dependent) ...... 14 4. Photosystem II...... 14 4.1. Structure ...... 14 4.2. Redox-active cofactors...... 15 5. The Oxygen-Evolving Complex ...... 18 5.1. Metal organization...... 18 5.2. The catalytic S-state cycle ...... 19 5.3. Protein ligands, channels and water binding ...... 21 6. The two redox-active tyrosines ...... 23 • • 6.1. The electronic structure of YZ and YD ...... 23 6.2 Oxidation and reduction...... 24 • 7. Metalloradical EPR signals I – Mn-YZ interaction...... 25 7.1. Background to the split EPR signals ...... 25 7.2. Inactive PSII ...... 26 7.3. Active PSII ...... 27 8. Metalloradical EPR signals II – Induction pathways ...... 32 • + 8.1. YZ formed via P680 ...... 32 • 8.2. YZ formed via the “NIR pathway” ...... 34 9. Metalloradical EPR signals III – Methanol and pH dependence of the OEC...... 39 9.1. Methanol...... 39 9.2 pH ...... 42 10. Conclusions and future perspectives ...... 51 10.1. Concerning YZ oxidation...... 51 10.2. Concerning the metalloradical EPR signals ...... 51 Acknowledgement ...... 52 Svensk sammanfattning ...... 55 References...... 58

Abbreviations

ATP adenosine triphosphate Car -carotene Chl chlorophyll DCMU 3-(3,4dichlorophenyl)-1,1-dimethylurea EPR electron paramagnetic resonance ESE electron spin echo ESEEM ESE envelope modulation EXAFS extended X-ray absorption fine structure FAD flavin adenine dinucleotide FNR FAD-containing ferredoxin-NADP+ reductase GAP glyceraldehydes-3-phosphate

HisZ histidine 190 on D1 HYSCORE hyperfine sublevel of correlation, a pulsed EPR technique LHC light harvesting complex MeOH methanol NADPH nicotinamide adenine dinucleotide phosphate NIR near-infrared (700-900 nm) OEC oxygen evolving complex PCET proton coupled electron transfer Pheo pheophytin PQ plastoquinone PSI photosystem I PSII photosystem II

QA primary quinone acceptor in PSII QB secondary quinone acceptor in PSII Ru5P ribulose-5-phosphate RuBisCO ribulose biphosphate carboxylase-oxygenase XANES X-ray absorption near-edge structure XES X-ray emission spectroscopy

YD tyrosine 160 on D2 YZ tyrosine 161 on D1

Introduction

In this thesis I will present new useful probes for one of the most fascinating enzyme mechanisms in the world, namely light-driven water oxidation. I be- lieve that the metalloradical EPR signals presented here will have great potential in the future to help elucidate the mechanism of the catalytic reaction.

1. Why do we study Photosystem II? At the dawn of life on Earth there was no oxygen in the atmosphere. Photo- trophic cyanobacteria evolved about three billion years ago, capable of using the energy of sunlight to extract electrons from the abundant supply of water and use these electrons for production of carbohydrates out of carbon dioxide. Oxygen was released as a by-product and started to accumulate in the atmosphere, thus changing the evolutionary pressure for all organisms. It is because of oxygenic photosynthesis that we, the aerobic life form, could evolve and populate the planet. Via respiration we utilize the oxygen released in the light reaction of photosynthesis to extract the energy stored in glucose, a product of the dark reaction of photosynthesis. Photosystem II is the enzyme that catalyzes the remarkable light-driven chemistry of water splitting.

1.1. We need energy Energy is the costs to be ordered in a universe that constantly strives to increase disorder. By breaking chemical bonds of molecules like sugar and the carbohydrates of gasoline, energy is released and this can be used for making new compounds in cells or when driving a car. Photosynthesis is the process by which energy from the sun is stored in chemical bonds. Mankind has al- ways thrived from this energy by eating plants to fuel her body or by burning firewood, oil, coal or natural gases to for example keep warm. In this way we can say that the energy used for all human activities today almost exclusively comes from the sun, but mostly in an indirect way via the use of fossil fuels (~80% of the global energy consumption comes from fossil fuels (1)). The oil and coal reserves we use today as fossil fuels were deposited during the age of the big fern forests 300-350 million years ago (Carboniferous), but these resources are limited and will eventually run out. The burning of fossil fuels for human activities, which on a geological time line can be con-

9 sidered instantaneous, releases most of the carbon dioxide that was fixed in photosynthesis of the fern forest a long time ago. The release of this potent greenhouse gas has led to the problem of global warming. Climate change and the limited nature of fossil resources are two good reasons why we need to use the unlimited energy from the sun (4.3 x 1020 J strike the earth every hour (2)) in a new way. The direct absorption of solar energy used in photo- voltaic, solar cells to create electricity is already in use, but electricity corre- sponds only to 17% (2006 (1)) of all energy used globally today. We also need a new fuel. The fuel would function as an energy carrier that can be stored and used outside the electrical grid.

1.2. Be biomimetic Can photosynthesis be useful in this context? We want to use solar energy and use an abundant raw material for fuel production. Why not be inspired by photosynthesis and mimic the light-driven water splitting catalyst of Na- ture, Photosystem II? The goal would be a manmade light-driven catalyst that could do the same chemistry as plants, algae and cyanobacteria, use light to oxidize water and yield an energy rich product, a fuel. Such a fuel could be H2, produced from recombining the electrons and protons generated in the water splitting catalysis. Hydrogen production occurs in nature in some species of cyanobacteria and algae. This reaction is catalyzed by a group of enzymes called hydrogenases. This second light driven catalyst could also be biomimetically inspired by the active site of the hydrogenase. Let’s try to make H2 from sunshine and water. This is the goal of the Swedish consortium of artificial photosynthesis (Figure 1).

2 H2O 2 H2 D 4 e- S 4 e- A

+ + O2 4H 4H

Figure 1. The catalytic scheme of light-driven oxidation of water and production of hydrogen. D = electron donor catalyst a PSII mimic, S = photosensitizer, A = electron acceptor catalyst a hydrogenase mimic.

Hydrogen is considered to be one of the energy carriers of the future, with the potential to replace oil. Big initiatives in both the EU and the USA have dedi- cated much research efforts to finding ways to produce and utilize hydrogen cheaply and efficiently. Large investments also been made to develop hydrogen fuel cells and infrastructure for the wide spread use of hydrogen as fuel.

10 The opportunity to contribute in the field of renewable energy is a strong motivation for increasing our understanding of one of the key enzymes in life, Photosystem II.

2. Oxygenic Photosynthesis In photosynthesis light energy is captured and stored in chemical bonds to be later used to drive cellular processes in the organism. Oxygenic photosynthesis occurs in higher plants, algae and cyanobacteria. It is the redox-process in which water is used as electron source in being oxidized, releasing oxygen into the atmosphere in the process, while carbon dioxide is reduced to carbohydrates. The overall reaction can be summarized by the general formula:

CO2 + H2O (CH2O) + O2 (1.1)

This process occurs in two steps. In the first step, light is captured to oxidize H2O (the light reactions). The electrons derived from water are in the second step used to reduce CO2 (the dark reactions).

2.1. Light reactions The light reactions take place in the thylakoid membrane of both cyanobacteria and photosynthetic eukaryotes. The thylakoid membrane is a fluid lipid bilayer in which all the proteins involved in the light reactions, antenna complexes, PSII, Cytb6f, PSI and ATP synthase are buried. The thylakoid membrane sits inside the chloroplast of eukaryotes and can be divided into the folded regions (grana stacks) and the single membrane parts (stroma lamella). The enclosed space inside the thylakoid is called lumen. In cyanobacteria the thylakoid membrane exists in a concentric multilayered structure within the cell directly. The light reactions can be divided into three phases: i) light absorption and energy delivery by chlorophyll antenna systems, ii) primary electron transfer in the reaction centers, and iii) secondary energy stabilization.

2.1.1. i) Light absorption and energy delivery Light is absorbed by many more chlorophylls than those that actually perform photochemistry. By funneling the absorbed light energy to the reaction center the existence of these so called antenna chlorophylls optimize the cellular cost of making the photosynthetic machinery against the available photon flux. Approximately 200 such chlorophylls are associated with each reaction center. From a rough estimation, a single chlorophyll molecule absorbs one photon every tenth of a second under full sunlight. Without the funneling efficiency of the antenna chlorophylls, the reaction center would remain idle in its unexcited state for most of the time without doing photochemistry.

11 The diversity in antenna complexes is large and they can be broadly divided into integral membrane antennas (such as the LHCI and LHCII of PSI and PSII) and peripheral antennas (such as the phycobilisomes of cyanobacteia).

2.1.2. ii) Primary electron transfer The conversion of the absorbed light energy to chemical energy takes place in reaction centers, which consists of large membrane-spanning protein complexes containing pigments and redox-active cofactors. The antenna chlorophylls funnel the trapped light energy to the reaction centers central chlorophylls (P) inducing an excited electronic state P*. The excited state P* is a potent reducing agent and will rapidly lose an electron to an electron acceptor (A) close by. This generates the charge separated state P+A- in a process known as charge separation. This primary state would easily recombine and lose the stored energy unless the even faster secondary processes + - take place to spatially separate the charges, D PA1A2 . In oxygenic photosynthesis there are two reaction centers, Photosystem II and Photosystem I, with their central chlorophylls called P680 and P700, respectively.

2.1.3. iii) Stabilization by secondary reactions In oxygenic photosynthesis the two photosystems (PSII and PSI) work in a linear (non-cyclic) electron transfer chain to secure the gained energy into stable intermediates (NADPH and ATP) that can be used in the catabolic processes of the dark reactions. The electron flow in the thylakoid membrane is illustrated in the Z-scheme (Figure 2) first described by Hill and Bendall (3).

–1.5 Photosystem II P700*

–1.0 A0 A1 P680* Fe-S Fe-Sx Fe-SA –0.5 Ph B Fd h FNR NADP+ (V)

m QA E 0.0 QB PQ h Cytb6f PC 0.5 P700

H O 2 OEC 1.0 Z P680 Photosystem I

Figure 2. The Z-scheme for electron transfer through redox-active intermediates in PSII and PSI. Starting from oxidation of water on the lumen side to the reduction of NADP+ on the stroma side of the thylakoid membrane. Two light-driven excitation events take place during this process.

12 The events in the electron transfer chain are summarized in Figure 2 and 3 and briefly described here starting at the point of the formed charge separated state, P680+Pheo-. After charge separation, P680+ is reduced by the redox active tyrosine Z that is re-reduced by electrons extracted from substrate water bound to the CaMn4-cluster in OEC, while the electron on the primary electron acceptor, pheophytin, is reducing QA and QB. After two charge separation events, the fully reduced and protonated electron carrier QBH2 leaves PSII and the electrons can then via the plastoquinone pool, en- ter the Cytb6f complex where plastocyanin is reduced. Plastocyanin, located at the luminal surface, transfers the electrons to PSI to reduce P700+. In PSI, excited P700* reduces the chlorophyll molecules in the primary electron acceptor A0 which further reduces the phylloquinones A1. The quinones transfer the photoexcited electron to three [4Fe-4S] clusters (FX, FA and FB). FB can reduce the [2Fe-2S]-containing ferredoxin (Fd), which in turn reduces NADP+ to NADPH at the enzyme FAD-containing ferredoxin-NADP+ reductase (FNR). In addition to the electron transport through protein complexes a proton gradient is built up across the thylakoid membrane (Figure 3). The protons released in water oxidation in PSII and the protons pumped into the lumen during the plastoquinone (PQ) /plastoquinol (PQH2) cycling by Cytb6f (two protons per electron transferred to PSI) are pumped back to the stroma by the enzyme ATP synthase during the synthesis of ATP.

Stroma 3H+ ATP 2 NADP 2H+ 2 NADPH ADP

8H+ e- FNR Fd e- PQ ATP PSII po Cytb6f PSI ol synthase e- e-

OEC PC 8H+ 3H+

+ 2 H2O O 4H 2 Lumen

Figure 3. The electron transfer chain through the thylakoid membrane described in the text. PSII – Photosystem II, OEC – oxygen evolving complex, PQ pool–pool of plastoquinones and plastoquinols, Cyb6f – Cytochrome b6f complex, PC – plastocyanin, PSI – Photosystem I, Fd – ferredoxin, FNR – FAD-containing ferredoxin- NADP+ reductase, blue dotted line – electron transfer, red dotted line – proton transfer, yellow bolts – light excitation of P680 (PSII) or P700 (PSI).

13 2.3. Dark reactions (not light-dependent) Both ATP and NADPH are used in the production of stable high energy molecules from carbon dioxide in the dark reactions of photosynthesis. The process by which CO2 is converted to carbohydrates is commonly known as the Cal- vin-Benson-Bassham cycle and can be considered as a two-stage process. In short, the production phase consists of the formation of glyceraldehydes-3- phosphate (GAP), the precursor of many carbohydrates, from ribulose-5- phosphate (Ru5P) and CO2. Energy is provided by the energy-rich ATP and NADPH. After this follows a recovery phase in which five molecules of GAP are shuffled to re-form the three starting Ru5P molecules. The enzyme cata- lyzing this important reaction is the ribulose biphosphate carboxylase- oxygenase (RuBisCO), which also may be the most abundant protein on earth.

4. Photosystem II Photosystem II, the light-driven water:plastquinone oxidoreductase is the enzyme of main interest in this thesis. It catalyzes the reaction:

light 2 H2O + 2 PQ 2 PQH2 + O2 (1.2)

Figure 4. The monomer of Photosystem II (2AXT – 3.0 Å resolution) colored by subunits; D1 - yellow, D2 - red, CP43 - green, CP47 - blue and smaller subunits in purple, turquoise, white and pink.

4.1. Structure PSII is a homodimeric protein-cofactor complex bound in the thylakoid membrane. It is 105 Å in depth, 205 Å in length and 110 Å in width, with a mass of ~700 kDa. At the moment the highest resolution crystal structures of the enzyme are solved at 3.0-2.9 Å resolution from Thermosynechococcus elongatus (4, 5). The PSII monomer is composed of the core heterodimer

14 subunit D1/D2, the chlorophyll containing subunits CP43 and CP47 and several smaller transmembrane subunits (Cytb559, PsbH, PsbI, PsbJ, PsbK, PsbL, PsbM, PsbN, PsbT, PsbX, PsbY, PsbZ, ycf12) as well as three extrin- sic subunits sitting on the lumen side (PsbO, PsbP and PSbQ in higher plants and PsbO, PsbU and PsbV in cyanobacteia) (Figure 4). The protein subunits hold in place the cofactors: 35 Chla molecules, 12 Car molecules, two pheo- phytins, three plastoquinones, two hemes (one in higher plants), 25 lipids, - 2+ bicarbonate, the CaMn4-cluster, Cl and one non-heme Fe (5).

4.2. Redox-active cofactors The electron transport through PSII can be followed by looking at the redox active cofactors (Figure 5), which are held by the core subunits.

Figure 5. Redox-active cofactors of PSII (3BZ1 – 2.9 Å resolution). Yellow ball – non heme iron coordinated by bicarbonate, blue ball – Cl- ion in the second coordination sphere of CaMn4 cluster (pink).

P680 is the collective name of the four central chlorophyll a molecules (ChlD1, PD1, PD2, ChlD2) that are the site of charge separation. Photon echo spectroscopy experiments point to the first step of excitation and electron transfer to be located on ChlD1, closest to PheoD1 (6). The cation radical + P680 , on the other hand, is most likely located on PD1, closest to YZ (7). The presence of oxidized YD has been shown by time-resolved fluorescence + measurements to localize the charge of P680 on PD1 (8). However, at low temperature it is speculated that the cation stays on ChlD1 (9). P680+ is one of the strongest oxidants in Nature, with estimated reduction potential of +1.1 +1.26 V (10-13). The key to the high oxidizing power of P680+ can probably be found in the protein matrix surrounding the chlorophylls. By reference to the location of the cofactors accepting or donating

15 electrons following a charge separation event, PSII is often referred to as having an acceptor side and a donor side.

4.2.1 The acceptor side – Pheo, QA, QB and the non-heme iron On the acceptor side the first electron acceptor is the pheophytin molecule on - D1 (PheoD1). The “working Em” of Pheo/Pheo has been found to be -525 to -420 mV (13, 14). Within ~2 ps after the formation of P680*, Pheo is reduced. - The anion radical Pheo that is formed quickly reduces QA (350 ps (15)), the first quinone acceptor, to stabilize the charge separated state. Only PSII centers with QA in the oxidized form, ready to receive the electron, can effi- - ciently trap the excited state of P680*. The midpoint potential of the QA/QA redox couple has been titrated many times under different experimental conditions (for a summary, see Fig 1 in ref (16)) and it has been found to be very sensitive to the measurement conditions (sample preparation, mediators and temperature). However, in active PSII centers at room temperature, the - Em of QA/QA is in the order of -80 mV (low potential form), while in inactive PSII (as well as in the frozen state), the Em of the redox couple is +65 - mV (high potential form) (16). QA must be oxidized to enable the second, third and fourth charge separation events to take place, which are essential for catalytic turnover at the OEC to evolve O2. The high potential form - “closes” PSII, making it harder for QA to reduce QB. Consequently the re- - + combination between QA and P680 /donor side takes place instead. This mechanism can work as protection against P680 triplet formation in PSII without the functional OEC (17). The occupancy of the QB site also affects - the Em of QA . It was observed that the presence of DCMU in the QB site - shifted the potential of QA/QA to -30 mV, while bromoxynil pushed the potential in the opposite direction to -130 mV (17). - In a process called the two-electron gate, reduced QB/QB is protonated to eventually form QBH2 after receiving 2 electrons and 2 protons, and is then able to leave the QB pocket. It enters the PQ pool in the thylakoid membrane - and is replaced by a new QB molecule. QA reduces the more loosely bound second quinone acceptor QB in the time frame of 100-200 μs or the semi- - quinone QB within 300-500 μs. These reduction steps involve structural changes observed as different positions of the quinone head group in dark compared to illuminated samples (18). Two channels for the Q/QH2 exchange were recently identified and they ended in the QB site and QC site. The head group of this newly found third quinone, QC is located between QB (17 Å away) and Cytb559 (15 Å away) (Figure 5) and QC can have an active role in the Q/QH2 exchange mechanism (5). The binding of QB to the QB site is easily out-competed by exogenous quinones and herbicides like DCMU or atrazine. The non-heme iron (Fe2+) is located between the two quinone acceptors (QA and QB), ligated by four histidines, two from D1 and two from D2. There is also the more flexible ligation of a bicarbonate ion that can either be bidentate (reduced Fe2+) or monodentate (oxidized Fe3+) in nature (19). The

16 redox potential of the Fe3+/Fe2+ couple is too high for be involved in the elec- - tron transfer from QA to QB, but exogenous electron acceptors like PPBQ in their semiquinone form can oxidize this iron (20). When this occurs, the 3+ - resulting Fe is rapidly reduced by QA (21). Bicarbonate is a labile ligand and varoius molecules (for example formate, NO, CN-) can competitively bind at the non-heme iron site and effect the QA to QB electron transfer rate. From FTIR studies it has been shown that the non-heme iron and bicarbonate are involved in an H-bond network to QB (22). One proposed role of the redox-active iron is protecting against active oxygen species (21).

4.2.2. The donor side – YZ, the CaMn4-cluster, the Car/Chl-Cytb559 side-path and YD On the donor side of PSII, there are several electron donating candidates to reduce P680+, but only one pathway leads to water oxidation. The direct primary pathway for water oxidation starts with the reduction of P680+ by the redox-active amino acid residue D1 Tyr161 called YZ. The kinetics of this reduction are multiphasic in the ns to μs time-scales (23). The distribution between the phases depends on the S-state of the CaMn4-cluster and will • be covered in chapter 6.2. The radical YZ thus formed is quickly re-reduced by the CaMn4-cluster. The kinetics of this process are also S-state dependent. The nature of the oxidation of YZ is the main focus of this thesis and will be discussed in the following chapters. The CaMn4-cluster is the catalytic site of water oxidation, composed of four manganese ions and one calcium ion (24). Two water molecules are oxidized one electron at the time in a stepwise manner to generate one molecule of oxygen in the last transition of a cycle called the S(0-4) state-cycle or Kok-cycle (25). During the catalytic turnover of the cycle, the four oxidizing equivalents are stored on the Mn ions (and maybe ligands). A simplistic description of the still much debated process of oxygen formation in the S- state cycle is that electrons and protons are removed from water until the O- O bond can form with direct release of O2. For a more detailed description of the catalytic S-state cycle see chapter 5.2. P680+ can under certain conditions be reduced by a secondary pathway of side-path electron donors: Cytb559, Chl-Z and Car. Car is the branch point + between Cytb559 and Chl-ZD2 (Figure 5) and it can directly reduce P680 (26). + Reduced Cytb559 will directly reduce Car , but in the presence of pre-oxidized Cytb559 Chl-Z will instead be the final electron donor. Oxidized Cytb559 can, in cyclic way, slowly be reduced again by electrons from the acceptor side of PSII. Since this reaction is both pH dependent and inhibited by DCMU, QBH2 ox is thought to be the electron donor to Cytb559 (27). However, a lower con- ox centration of DCMU is required to inhibit Cytb559 re-reduction than to inhibit the QB reduction (27). This suggests that the newly found QC site (5, 28), close to Cytb559, can be involved in this cyclic side-pathway. It is debated whether the role of this side-pathway is photo-protective in nature by remov-

17 ing long-lived P680+ radicals, as first proposed, or to reduce Car+ to allow Car to act as a singlet O2 quencher as was proposed in (29). However, experiments on an over-reduced PQ pool found in a tobacco Cytb559 mutant, has suggested that the function of Cytb559 is to keep the PQ pool and the acceptor side oxidized in the dark (30). Recent work in oxygen-evolving PSII has shown that the multiple Chls in the light harvesting subunits CP43 and CP47 could also reduce P680+ at cryogenic temperatures (31). The second redox-active tyrosine in PSII, YD (D2 Tyr160), is homologously located on the D2 subunit. In its reduced state, it can also function as an electron donor to P680+. This property is very important in the photo- • assembly of the CaMn4-cluster, which is accelerated in the presence of YD + (32). The oxidized YD is proposed to tune the redox-potential of P680 by localazing the cation on PD1 (8). YD can also function as an electron relay for the decay of the S-states in the dark, where the CaMn4-cluster can be either • oxidized or reduced by YD and YD, respectively (33).

5. The Oxygen-Evolving Complex The catalytic site of water splitting is the oxygen-evolving complex (OEC), also called the water oxidizing complex. The OEC is composed of a metal cluster assembled from four Mn and one Ca2+ bridged by oxygens, Cl- (required for PSII activity), substrate water, and the ligand sphere consisting of amino acid side chains of mostly D1 protein residues including YZ.

5.1. Metal organization The question of exact organization of the metal ions in the cluster is not yet resolved, but good estimates are present in literature from the combination of X-ray diffraction, X-ray spectroscopy and EPR spectroscopy data. 2+ The assignment of Ca as part of the oxygen-bridged Mn4Ox-cluster has been confirmed by the X-ray crystallography structure (4, 5, 24). However, the resolutions of the refined structural models are still low ( 2.9Å). fur- thermore, the cluster could be over-reduced and maybe lose structural motifs due to radiation damage during X-ray exposure (34, 35). Therefore, the reli- ability of the exact geometry of the cluster as presented in the crystal structure, especially with respect to interionic distances, has been called into question. Polarized EXAFS studies, while reducing the X-ray dose to which the sample is exposed to, on PSII crystals, a technique that gives metal-metal and metal-ligand distances, has produced arguably better data for the distances between the metal ions and their internal organization (36). Three out of all possible geometrical arrangements of the metal ions gave calculated EXAFS spectra that gave good fits to the polarized experimental data. These models, which correspond to the S1-state, were placed into the ligand sphere

18 of the 3.0 Å resolution crystal structure (4, 36) (Figure 6A). As the ligand sphere from the crystal structure was not relaxed after insertion of the EXAFS-derived model of the CaMn4-cluster, the position of the amino acids closest to the cluster were not in optimal position to be ligands. To overcome this problem QM/MM and DFT calculations have been used to optimize the cluster together with ligands, whilst attempting to maintain a good fit to the experimental EXAFS data (37, 38).

Figure 6. (A) One structural model (Model II) of the CaMn4-cluster from polarized EXAFS spectroscopy (36) placed in the protein ligand framework from the 3.0 Å resolution PSII structure (4). (B) Suggested oxidation states of and calculated ex- -1 change coupling constants (cm ) between the Mn-ions in the S0- and S2-states from reference (39).

5.2. The catalytic S-state cycle The reaction catalyzed by the OEC is the oxidation of water with evolution of molecular oxygen:

+ - 2 H2O O2 + 4 H + 4 e (1.3)

To evolve one molecule of oxygen, the absorption of four photons drives four charge separation events that extract four electrons from the CaMn4- cluster, leading to accumulation of the oxidizing equivalents. The oxidation of water leads to re-reduction of the CaMn4-cluster back to its lowest oxidation state. Therefore water is ultimately the source of electrons in oxygenic photosyntheis. Kok and coworkers (25) suggested a four step mechanism (Snn+1) called the Kok-cycle or S-state cycle to explain the oscillating pattern of flash-induced oxygen release from chloroplasts,. The CaMn4-cluster • is oxidized in steps by photogenerated YZ starting in the S0-state, the most reduced state, followed by to the S1-state, the dark stable state. This state is further oxidized to the S2- and S3-states, the metastable states that advance to the transient S4-state, at which point substrate water bound to the CaMn4- cluster is oxidized, with the concomitant release of oxygen and the return of

19 the cluster to the S0-state. Recently, an extended S-cycle was proposed, in- corporating the sequential and alternate removal of the four electrons and the four protons from the CaMn4-cluster in the total of eight steps leading up to the formation of oxygen (40). To investigate the electronic structure of the CaMn4-cluster in different S- states, both EPR spectroscopy and XAS have been used. 55Mn-ENDOR spectroscopy has been successfully applied in the paramagnetic S2- and S0- states (39, 41). In combination with other spectroscopic techniques, the oxidation states and exchange coupling of the individual Mn ions have been proposed in the S0- and S2-states (39) (Figure 6B). With XAS measurements, all S-state transitions have been addressed (42) both at cryogenic temperature and room temperature (43). The structural and electronic features in each transition were found to be independent of temperature. The following structural and electronic changes have been proposed with respect to the individual S-state transitions (assuming the high III,III,III,IV valence cluster Mn in the S0-state (39)): S0S1 Mn-centered oxidation occurs during this transition, most likely from MnIII to MnIV accompanied by a shortening of a Mn-Mn distance (2.8 Å to 2.7 Å) that has been assigned to the deprotonation of a μ-hydroxo-bridge (43). III IV S1S2 Only Mn-centered oxidation, Mn to Mn , takes place during this transition without changes in the bridging mode (43). S2S3 This is the most debated of the transitions, with split opinions concerning the nature of the oxidation. The oxidation is accompanied with significant structural changes (43-45). Cur- rently, the main point of contention is whether the one- electron oxidation is Mn-centered (25, 46, 47) or ligand-based (42, 48) in nature. Based on interpretation of XANES data, advocates of the Mn-centered oxidation hypothesis assign the step to the oxidation of a five-coordinate MnIII ion into a six- coordinate MnIV (46, 49). This is accompanied by the shortening of the Mn-Mn distance from >3.0 Å to 2.7 Å, which is consistent with the formation of a third di-μ-oxo-bridge with a proton being expelled (43). Recent DFT calculations, have also suggested structural changes in the S2S3 transition that involve the binding of a water molecule with subsequent removal of a proton and Mn oxidation (50). Supporters of the ligand-centered oxidation hypothesis argue for the formation of a Mn-μ-oxo-radical in the S3-state. The EXAFS data interpreted instead as showing the lengthen- ing of a Mn-Mn distance from 2.7 Å to ~3.0 Å. They also relay on the lack of difference in the K XES difference spectra (compared to S0-S1 and S1-S2) to exclude the possibility of Mn oxidation (48).

20 S3S0 The last S3-[S4]-S0 transition involves several steps to complete the S-state cycle, including one oxidation (to reach S4), formation and release of O2, release of two protons and re- binding of substrate water molecule(s). The first step in this chain of events leading to O2 formation was observed to be a deprotonation of the cluster, presumably via CP43-Arg357 (51). Mutation of the arginine residue severely impairs catalytic activity (52). Several models exist for this last elusive mechanistic step of water oxidation (50, 53-55). The models can be classified according to the mode of O-O bond formation (54): (i) several variations of oxyl radical reactions, (ii) coupling of bridging oxo ligands, and (iii) attack of terminally bound water/hydroxide upon MnV=O.

5.3. Protein ligands, channels and water binding

The structure of the CaMn4-cluster is supported by the protein ligand framework. The first coordination sphere of ligands includes the carboxylic groups of the amino acids D1 Asp170, D1 Glu189, D1 Glu333, D1 Asp342, D1 Ala344 and CP43 Glu354, as well as the imidazole ring of D1 His332 (Fig- ure 7, left panel). Most of these amino acids have been mutated and studied by FTIR spectroscopy (56). From these studies it has been proposed that the C-terminal of D1, namely Ala344, coordinates the Mn ion that changes oxidation state during the S1S2 transition (57). From FTIR studies, D1 Asp170 (56), D1 Glu189 (58) and D1 Asp342 (59) do not seem to coordinate to Mn ions that are oxidized during the S0S1 or the S2S3 transitions. However, mutation of CP43 Glu354 significantly lowers the O2 evolution yield and potentially inhibits advancement above the S3-state (60). From thermoluminescence measurements, it was concluded that mutation of the histidine ligand D1 His332 blocked the S-cycle at the S2-state (61). Both His332 and Glu333 mutants seem to diminish the affinity for Ca2+(62) although as evident from the crystal structure models they are not ligands to Ca2+ (4, 5) (Figure 7, left panel). Both D1 His332 and D1 Glu333 remain to be investigated with FTIR, which will hopefully disclose the amino acid ligating the Mn oxidized during the S0S1 transition. The picture is cur- rently not coherent between the ligands found in the crystal structure and those proposed by mutation studies, and a lot of work remains to be done. Many of the amino acids in the second coordination sphere form a hydrogen bond network around the cluster ligands. This includes the strictly con- served CP43-357 (dark blue, Figure 7) that is vital for the catalytic function (52) and believed to be involved in an important deprotonation events during the higher S-state transitions (51, 63). Also, the essential Cl- cofactor has recently been found to be positioned at the entrance of the putative proton channels (5) (Figure 7, turquoise).

21 EF B Large Channel Channel (ii)

1 G Back Channel Ca F Cl- Channel (i) 4

Broad Channel D C Channel (iii)

A1 CD A2

Figure 7. Left panel, the first coordination sphere of protein ligands around the - CaMn4-cluster (C-grey, O-red, N-blue, metal-orange) with Cl (turquoise) and CP43- Arg357 (dark blue) all marked in relation to YZ (right corner) (PDB file 3BZ1). Right panel, the proton (yellow), water (red or blue) and oxygen (red or blue) channels around the CaMn4-cluster. The letters refer to the assigned channels in (5) and in italic the corresponding channels in (64, 65).

Channels for proton release, entry of substrate water and release of O2 have been computed (5, 64, 65) from the crystal structures (4, 5, 24) (Figure 7, right panel). All together eight channels, connecting the CaMn4-cluster with the lumen, were found in the latest 2.9 Å resolution structure. The most hy- drophilic channels (Figure 7, right –yellow), especially the channel called C/broad channel/channel (ii), have most likely the function of proton release, where the protons are attracted by the negative chloride ion (turquoise) placed at the proximal end of the channel. The remaining channels (blue and red) are wide enough (2.6-2.8 Å) to transport either water or oxygen to and from the catalytic site, respectively (5). The function of the water channel(s) could be to control the delivery of substrate to the active site (66). In (64) a possible control gate in the form of a gap between two channel systems was observed, involving Ca2+ and the mechanistic important D1 residues, Tyr161 (YZ), His190 (HisZ), Glu189, Phe186 and Asn165. This gate could potentially be involved in regulating the water access to the substrate site to avoid for example side reactions caused by “flooding” (reviewed in (67)). To investigate the pathway for molecular oxygen transfer, PSII was crys- tallized with Xe under pressure (5). In the most recent study, none of the nine Xe sites identified were localized in the proposed channels; instead the Xe sites were located in the hydrophobic environment of lipids and fatty acids mostly in the membranes spanning part of the protein. Potentially, one function of the many lipids found within the membrane part of PSII could be to guide the released O2 to the stromal side of the thylakoid and shield the redox cofactors from O2.

22 Water binding has been addressed both kinetically and structurally in the different S-states. From time-resolved MIMS measurements, both fast and slow water exchange rates were observed correlated to two independent water binding sites (68). Later, kinetics for the slow exchange site were observed in all S-states, while kinetics for the fast exchange site only were found in the S2- and S3-states (69-71). From FTIR studies on PSII it was found that the S2S3 and S3S0 transitions were sensitive to high levels of dehydration suggesting insertion of substrate in these transitions (72), which 16 18 was also the conclusion based on D2 O/D2 O FTIR spectra (73). Only one 17 17 O (water) was coupled to Mn in a recent O-HYSCORE study of the S2- state (74), indicating the presence of one Mn-bound water in the S2-state.

6. The two redox-active tyrosines

The two tyrosine residues, YZ (Y161) and YD (Y160) are homologously located on the D1 and D2 subunits, respectively. They are both redox-active in donating an electron to P680+, but functionally very different due to the presence of the CaMn4-cluster on the D1 subunit. On the D1-side is the tyro- • syl radical, YZ , that is formed after charge separation quickly re-reduced by • the CaMn4-cluster, wheras YD on the D2-side can remain oxidized for hours.

• • 6.1. The electronic structure of YZ and YD EPR spectroscopy has been very useful in determining the electronic structure of the tyrosine radicals. The first EPR signal discovered from PSII by Commoner et al in 1956 was Signal II (75). Much later could Signal II be • assigned by Barry & Babcock to originate from the tyrosine radical, YD , based on experiments using cyanobacteria grown on isotropically labeled • tyrosine (76). The characteristic lineshape of the X-band spectrum of the YD radical (Figure 8B) results from g-anisotropy and hyperfine couplings to the -ring and -methylene hydrogens. Using high-field EPR (245 GHz), the anisotropic g-values have been resolved for several tyrosine radicals in different protein environment. It was shown that the gy (~2.0045) and gz (~2.0021) components are insensitive to the protein environments, in contrast to the gx component (~2.0064 – 2.0090) that varied distinctively and was indicative of the degree of hydrogen-bonding of the phenol proton (77, • • 78). For both YD and YZ , the hydrogen hyperfine couplings have been measured and the unpaired spin densities of the phenoxyl-ring carbons (numbered in Figure 8A) have been calculated (79-82). Large couplings from the ring hydrogens 3 and 5 reflect the high spin density on these carbons (~0.25 on each), while small couplings on the hydrogens 2 and 6 ring relate to the low spin density of -0.06 on their carbons. The largest spin density (0.37) rests on the ring carbon C1, and projects to the -methylene hy-

23 drogens in a way depending on the angles between the hydrogens and the - ring plane. These angles, determined by the ring orientation, can vary con- siderably among different tyrosine radicals. Since at least one of the - methylene protons often feels the strongest coupling of the spin of the tyrosine radical these angles are also reflected in the line shape of the EPR spectrum. Approximately 0.28 spin resides on the oxygen and the -0.03 spin on the ring carbon C4.

O A B 4 H 5 H 3

6 H 2 H 1

H C H 2 1

CH 3320 3340 3360 3380 Magnetic Field (G) N C Figure 8. In A, schematic tyrosine radical with the -ring carbons and -methylene hydrogens numbered. In B, the continuous-wave X-band EPR spectrum of the stable • YD radical measured at 15 K, microwave power 1 μW, modulation amplitude 3.2 G.

6.2 Oxidation and reduction

The reduced forms of both YZ (83, 84) and YD (85, 86) are protonated (87). The oxidation of YZ (D) is coupled to deprotonation of the phenol (YZ (D)-OH) • with the formation of the neutral radical YZ (D)-O . This is supported by the g- value 2.0046 of the radical EPR spectrum of YZ and YD closer to the value expected for the neutral radical (2.0044) than the protonated radical (2.0032) • (76). The oxidation of YZ-OH to YZ-OH would also be thermodynamically unfavorable due to the estimated high reduction potential of this redox couple (~1.4 V) (88), which is well above the potential of P680+/P680. The • • presence of a hydrogen bond to the phenolic oxygens of both YZ and YD neutral radicals was evident from a 2H-ESE-ENDOR investigation (89). Also studies of the sensitive gx-component as determined by high field EPR, has indicated the presence of hydrogen bonding to both tyrosine radicals (78). • However, the YZ radical in Mn-depleted PSII has a broader gx peak, which has been interpreted as a distribution arisen from disorder in the protein environment. From the protein crystal structure (4, 24) and site-directed mutagenesis work (89-92), it is evident that the two homologously located histidines, D1 His190 and D2 His189, are likely hydrogen bonding partners/proton acceptors to YZ and YD, respectively.

24 + The multiphasic reduction of P680 by YZ (oxidation of YZ) is S-state dependent (23, 93). The major kinetic component in S0 and S1 is the fast ns- kinetics (t½ 20-60 ns), while the major component in S2 and S3 is the slow ns-kinetics (t½ 250-300 ns). There exists also an S-state dependent minor component in μs time range (94). These different kinetic phases display different kinetic isotope effects: The fast ns-kinetics show virtually no H/D isotope effect and small activation energy (10 kJ/mol) (93, 95-97), while the μs-phase, shows a significant H/D isotope effect (97). At physiological pH, is the oxidation of YZ-OH coupled to deprotonation of the phenol proton. The kinetic isotope effects on the multiphasic reduction kinetics of P680+ can be interpreted in the context of the fate of this phenol proton. Therefore, the ns-kinetics time range is suggested to reflect the deprotonation of YZ-OH in a through proton shift in a well-tuned hydrogen bridge, while in the μs time range is suggested to reflect deprotonation that is couple to internal proton movements in a H-bond network (97, 98). • The radical YZ is quickly reduced by the CaMn4-cluster. To facilitate this fast re-reduction, the transferred proton probably remains hydrogen bonded to the tyrosine to facilitate a rapid shift back to the phenolic oxygen. The rate • of YZ reduction is S-state dependent and has been reported with life times of < 3–250 μs for S0S1, 30–110 μs for S1S2, 100–600 μs for S2S3 and 1– 4.5 ms for S3S0 ((99) and references therein). These rates show different kinetic isotope effects and activation energies (96). The S2S3 transition, has the largest kinetic isotope effect and also most pronounced pH dependence on the electron transfer rate (96).

• 7. Metalloradical EPR signals I – Mn-YZ interaction + YZ, the primary donor to P680 , is difficult to study because of the fast oxidation and reduction in intact PSII. Therefore, YZ has mostly been studied in Mn-depleted PSII, where the radical has a longer lifetime, 150-700 ms (100) compared to 0.03-1 ms in intact material (101). However, these studies are of limited value to understand the molecular details of YZ in the intact sys- • tem. New probes for the functional structure of YZ are the metalloradical EPR signals from the OEC (split EPR signals). These EPR signals are the focus for this thesis.

7.1. Background to the split EPR signals Electron paramagnetic resonance spectroscopy (EPR) is a technique used to detect species with unpaired electrons (spins) like free radicals and transition metal ions. PSII is crowded with EPR active species and provides a heaven for EPR spectroscopists. After charge separation, an electron is transferred between the redox-active cofactors through the enzyme, and is detectable as

25 the radical species it creates. Both Mn, present in the catalytic site, and Fe, present in the heme of Cytb559 (in cyanobacteria also Cytc550) and the non- heme iron on acceptor side, are also EPR active species in certain oxidation states. This makes EPR spectroscopy an excellent technique to probe the catalytic activity of PSII. If two paramagnetic species are present simultaneously at a short distance (5-10 Å), their spins will magnetically interact and a new set of EPR lines with a common center will appear at the expense of the individual lines of the two magnetically interacting species. The nature of this interaction can be (1) between the magnetic dipoles of the electrons (dipolar) and (2) a spin exchange interaction. In (1) the dipole’s magnetic field senses the orientation of the neighboring paramagnetic dipole(s) and this can be observed in the EPR spectrum as a line broadening. In (2) the interaction exceeds the pure dipolar on shorter distances (r < 5Å) and is due to the electrical and orbital moments’ overlap. Only local fields that remain “constant” compared to the spin-spin relaxation time (T2) are effective in shifting the resonance frequency of a given spin (102). The shape, line width, relaxation properties and g-value provide information about the interacting species and the interaction itself.

7.2. Inactive PSII A metalloradical EPR signal from OEC was first found in PSII with Ca2+ depleted from the Mn4-cluster and therefore incapable of O2 evolution. The split EPR signal is formed by illumination at 0 °C, giving rise to a signal 164 G wide signal centered around g = 2 (103) (Figure 9, white – S2). During the nineties, many researchers discovered that various treatments that removed or displaced Ca2+ or Cl- from OEC generated the same type of split EPR signals upon illumination at 0 °C, although the width of the signal varied with treatment (104-117). These split EPR signals arise from a radical (S = 2+ ½) in magnetic interaction with the (Ca /)Mn4-cluster in a modified S2-state (113, 118) (first thought to be a formal S3-state). The identity of the radical was first suggested to be an oxidized histidine (107, 119). However, the most • • favored candidate is YZ , based on EPR studies correlating YZ with the split signal (109, 116, 120) and strong support from ESE-ENDOR and ESEEM • studies (121, 122) attributing the signal to S2YZ . Unlike the interpretation above of the split signal in Ca2+-depleted PSII, one research group interpret the radical species involved in the split signal differently. They found the light induced EPR spectrum to be composed of two overlapping signals: a symmetric doublet and an asymmetric singlet-like signal (123). The origin of the doublet signal was proposed to involve a di- • polar interaction between YZ and another organic radical nearby (5.3 Å) (124, 125) recently speculated to be D1His190 (126). The origin of the singlet-like signal was proposed to involve the interaction of the Ca/Mn4-

26 cluster with a nearby radical X• (124). The same radical X• was speculated to be involved in a different kind of split signal (g = 2 broad signal) generated by illumination at slightly lower temperature (243 K) also in Ca2+-depleted PSII (“S1-state”) in the presence of DCMU (126) (Figure 9, white). The argument for not assigning radical X to YZ in (126) were based on distance: the distance between the interacting spins in the singlet was estimated to be at most 6.3 Å which was considered to be short for YZ to be involved. How- ever if we consider the distance between YZ and the CaMn4-cluster in the recent crystal structure (4), the 6.3 Å distance could as well argue for radical 2+ X to be YZ, especially in the Ca depleted PSII, where the Mn-ions and the YZ radical can come even closer to each other.

2+ S0 - intact PSII, vis S1 – intact PSII, vis S1 –Ca depleted

2+ S2 –Ca depleted

3200 3300 3400 3500 Magnetic field (G) 3200 3300 3400 Magnetic field (G) S2 – intact PSII, IR

H+ e- e- S1 S2 – intact PSII, 190-77 K S S 2 3200 3300 3400 2 H2O 0 Magnetic field (G) H+ S – intact PSII, IR O e- 3 2 [S4] S3 H+ e- H+

3200 3300 3400 3500 Magnetic field (G)

Figure 9. Summary of S-state dependent split EPR signals induced in PSII under different experimental conditions (e.g. intactness, temperature, light quality). Spectra 2+ 2+ adapted; S1 – Ca depleted from Mino & Itoh 2005 (126); S2 – Ca depleted from Boussac et al 1989 (103); S2 – intact PSII, 190-77 K from Ioannidis et al 2006 (127).

7.3. Active PSII At the start of the new millennium, split EPR signals of a metalloradical nature induced in intact PSII were discovered (127-131). They can all be induced by illumination at cryogenic temperatures and found to oscillate with the S-states (summarized in Table 1, Figure 9). For the first time, were electron transfer intermediates trapped in active PSII. I will here prove them

27 useful to probe both YZ oxidation and the S-states of CaMn4-cluster. Results from Paper I-VIII will be reviewed in the following chapters together with the relevant literature. However, it is clear that this work is only starting and there are less than 20 publications about these signals compared to the S2 multiline signal, which is studied or used in more than 300 publications (ISI web of science search).

7.3.1. Assignment In order for the split signals to be useful probes, the identity of the interacting species involved must be confirmed. A number of observations suggest that the interaction signals have their origins in metalloradical species: the split EPR signal is in general wider than expected for an organic radical (from the splitting of the radical spectrum), the microwave power needed to half saturate the EPR signal (P½) is high in the mW-range (a non-interacting organic radical is expected to saturate in the μW-range) (Table 1), and the split EPR signal spectrum narrows into a radical spectrum at increasingly higher temperatures (132). The assignment of the metal species being the CaMn4-cluster is based on the S-state dependence of the signals ((130), Pa- per I and V) and the similar effect of methanol on the induction and shape of the split signals and on other of the S-states characteristic EPR signals (Pa- per II). The simplest proof of the CaMn4-cluster nature of the split EPR signals, might be that the intact cluster is actually required for the metalloradical signals to be induced at cryogenic temperatures. The total spin of the CaMn4-cluster is S-state dependent, suggested to be non-integral spin in the S0- and S2-states, S = 1/2, 3/2, 5/2.., and integral spin in the S1- and S3-states, S = 0, 1, 2, 3.., based on the paramagnetic EPR signals observed in the different S-states (133-136). In addition to the nature of the spin-spin interaction, these differences in spin states are contributing to the different spectral shapes of the S-state dependent split signals. • The assignment of the interacting S = ½ radical to YZ is based on the following: 1. The similarity of the split signals in active PSII to the split EPR signals • in inactive PSII, that is identified to involve YZ (121, 122). 2. Focusing on the radical region of the light-induced EPR spectrum (g = 2 measured with low modulation amplitude to increase the resolution of narrow signals) reveals a fast-relaxing tyrosine like spectrum (~20 G, g~2.0046) with a fast decay kinetics (Paper I, Paper VII, (137, 138)). 3. All the split signal spectra collapse into a tyrosine radical spectrum at temperatures >100 K (132). 4. The distance between YZ and the CaMn4-cluster is suitable for a paramagnetic interaction signal, whilst the known secondary electron donors + to P680 (Chl, Car and Cytb559) are to far away from the CaMn4-cluster to be involved.

28 • 5. The assignment of YZ in the S1-state is strongly supported in our preliminary pulsed EPR study of the Split S1 EPR signal (139). We observed a light induced tyrosine radical in the ESE field sweep spectrum (Figure 10A) at 3.8 K in intact PSII in the S1-state. From the ESEEM measurements at the Split S1 signal peak (g = 2.035), we found indica- tions that the unpaired electron spin could be coupled to one of the - methylene protons of the tyrosine. The coupling to the -methylene proton was observed as the light induced peak at 1.9 MHz (140) in the two- pulse ESEEM frequency domain (Figure 10D). In the three-pulse measurements (Figure 10E), a light-induced peak appeared at 3.8 MHz, which possibly reflect spin coupling to a nitrogen. A similar peak was observed in the ESEEM frequency domain of a 4(5)-CH3-imidazole radical (141). However, since a spin-spin interaction signal was investi- • gated the observed nitrogen coupling can reflect either the YZ radical or nitrogen ligation to the CaMn4-cluster. Additionally, in previous studies • of YZ in Mn-depleted PSII, (81) no nitrogen coupling similar to nitrogen • coupling observed in the case of YD could be found (142). This sug- • gested disorder in the hydrogen bonding of YZ in the depleted system. • Our results could be the first observation of nitrogen coupling to YZ , which would be indicative of a more ordered hydrogen bonding in intact PSII compared to Mn-depleted PSII.

The only published simulations of the split signals in intact PSII are the X- band and W-band spectra of Split S1 EPR signal, that can be taken as a test of the metalloradical nature of the split signal. It has been simulated assuming a total spin S = 0 (ground state) and S = 1 from the low lying first excited state of the CaMn4-cluster in the S1-state in a weak exchange interaction with a S = ½ radical (143).

Figure 10. PSII ,with reduced YD and poised in the S1-state, were measured in the dark or during continuous illumination at 4 K or after 25 min dark decay also at 4 K. The two pulse ESE field sweep spectra of the radical region (A) and split region (B). Evaluation of data in A: deconv 1 represents light induced stable signal and deconv • 2 represents the decaying part of the light induced signal compared to YD (white line). The arrow in B indicates the position for the ESEEM measurements. Two- pulse (C, D) and three-pulse (E, F) ESEEM measurements in the time domain (C, E) and corresponding Fourier transforms (D, F). EPR conditions (ESE field sweep): A – microwave power 0.5 W, B – microwavepower 25 mW; microwave frequency 9.71 GHz and temperature 4 K, the /2-- pulse sequence was 16-200-32 ns, with repetition rate of 200 s. EPR conditions (ESEEM): C, D – /2 pulse 16 ns, starting 100 ns, increment 4 ns, repetition rate 400 s; E, F – /2 pulse 16 ns, 160 ns, starting T 100 ns, increment 4 ns, repetition rate 600 s, phase cycling. Microwave frequency 9.71 GHz, field position 3407 G and temperature 3.8 K

30 Table 1. Properties of the split EPR signals induced by cryogenic temperature illumination in active PSII. S-state of induction induction yield dark decay EPR signal P½ (mW) +MeOH, [MeOH]½ optimum pH induction light temp. (% PSII) t½ at 5 K (min)

Split S0 a b c d c • S0 400-690 nm 4-15 K ~50 3 >10 (5 K) new signal 0.54% 4.87.5 (S0YZ )

Split S1 a b b d e • S1 400-690 nm 4-15 K ~40 3 >1 (5 K) no signal 0.12% 4.7

Split S2 f f f f f • S2 visible 77-190 K - few min >20 (10 K) inducible at 10 K - (S2YZ )

“Split S2” g h g g • S2 visible, NIR 4-15 K <10 stable >30 (10 K) - - (S1YZ ) stable (NIR) a “Split S3” a h b d 4.87.5 • S 400-900 nm 4-50 K ~50 3, 15 & stable >5 (5 K) no signal 0.57% (S Y ) 3 (NIR) h 2 Z (vis) b

broad g = 2 h h h h • S2, S3, S0 visible 4-15 K - 3 >10 (10 K) - pH>7.8 (SxR ) a Paper IV b Paper I c Paper VII d Paper II e Paper III f Ioannidis et al 2006 ref (127) g Paper VIII h Unpublished

8. Metalloradical EPR signals II – Induction pathways The induction conditions (temperature and light) vary between the different S-state dependent split EPR signals (summarized in Table 1). Two types of light regimes were identified; 400-690 nm (visible) and 700-900 nm (NIR) (Paper IV). Based on the wavelength of the inducing light, conclusions can be drawn on the light absorbing species behind the signal formation. The split EPR signals will be divided in two categories here: the split signals induced as a result of charge separation centered on P680 (Figure 9, green), and split signals induced by Mn excitation (Figure 9, pink).

• + 8.1. YZ formed via P680 8.1.1. Electron (and proton) transfer at cryogenic temperatures To understand the origin of the split EPR signals in intact PSII we first need to consider the electron transfer events after charge separation at cryogenic temperature. The electron can reduce QA, but further transfer to QB is blocked by the low temperature since this would require structural changes + and protons movement. P680 can be reduced by YZ or secondary donors red • (Cytb559, Car/Chl, YD ). If YZ is formed, further reduction by the CaMn4- cluster is restricted by the low temperature, since this would also require - • extensive movements. Instead, QA and YZ will recombine with a t½ ~3 min at 5 K (Paper I). Protons can nevertheless tunnel at liquid helium temperatures if the system is in a “tunneling-ready” configuration with a short proton-transfer distance (144). However, environmentally coupled H-tunneling models, where protein dynamics is coupled to H-transfer chemistry via so called promoting vibrations predicts that the dynamics of H-transfer could freeze out at cryogenic •+ temperatures (145). This was observed when YD was trapped at 1.8 K and appeared from the high-field EPR spectrum to still be protonated while it was by contrast deprotonated when it was trapped at 77 K (146). These results would be the prediction if H-transfer is coupled to dynamic processes. However, this is probably not the case for YZ in intact PSII (at physiological temperature) where the almost activationless fast electron transfer to P680+ and small kinetic isotope effect argue for a very short H-transfer distance and a proton transfer that is not rate-limiting. The idea would be that electron transfer from YZ at cryogenic temperature can only happen if the tyrosine-OH is frozen in the “tunneling-ready” configuration, i.e H-bonded to the N on HisZ supported by a set H-bond network in OEC. For support of this idea see next chapter 9.2.

8.1.2. The split EPR signals from S0-, S1- and S2-states The split EPR signals in this category can be induced by monochromatic light in the visible range (400-690 nm) (Paper IV) and show a clear correlation to

32 - the induction of QA (Paper VI, (127, 147)). Based on these observations, light absorption leading to P680 centered charge separation is the most likely ori- • gin of the radical species (YZ ) involved in the S0-, S1- and S2-state split EPR signals. The EPR spectra, recorded at liquid helium temperatures, represent • frozen snapshots of water oxidation intermediates (SnYZ ). In the S0- and S1- states, the split EPR signals are induced by illumination at 4-15 K, while the induction of the S2-state signal requires higher temperatures to overcome a thermal barrier (77-190 K) (127), but see Paper VIII. The thermal barrier is speculated to be a secondary proton transfer from the base partner (HisZ). In Paper I the shape of the Split S0 EPR signal was deduced from its decay associated spectrum (t½ = 3 min). It is composed of a 140 G symmetrical split and a 20 G sharp derivative signal both centered at 3355 G (Figure 9, green – S0). The relaxation properties measured as the microwave power of half saturation of the EPR signal (P½) show a fast relaxation behavior (>10 mW) that is indicative of transition metals (Paper I and VII). The yield of this signal is ~50%, estimated from the induction and decay of the Split S0 signal - and QA in the same sample (130, 147). The yield of the 20 G fast decaying and fast relaxing (0.3 mW) radical at g = 2.0041 in the S0-state is ~5% of PSII • centers (Paper VII). The small radical probably represents non-interacting YZ in PSII centers with spins populating a very fast relaxing (higher) spin state of the CaMn4-cluster that on average will not be felt by the radical. The symmetric EPR spectrum of Split S0 signal represents the intermedi- • ate S0YZ and is the first visible EPR signal from the S0-state without methanol in higher plants (with the exception of the S0 multiline signal in Thermo- synechoccus elongatus (148)). The Split S0 signal is a very useful probe of the S0-state in deconvoluting the S-states in laser-flashed PSII samples, where the S-states mix due to misses (Paper V). In Paper I, the shape of the Split S1 EPR signal was confirmed by analysis of the decay associated spectrum (t½ = 3 min) with a peak at g = 2.0035, a shoulder on the high field side and a 20 G radical centered at 3355 G. Both the relaxation properties (>1 mW) and yield (~40%) of the S1-state split EPR signal are similar to the S0-state signal (Paper I, (147)). In Figure 11, we analyzed the radical region of the Split S1 signal. The 20 G light induced radical in the S1-state was fast decaying, fast relaxing (P½ ~0.5 mW) and produced in ~16 % of PSII centers under long illumination. The fast relaxation of the 20 G radical increased the resolution of the tyrosine hyperfine • structure compared to the spectrum of YD (Figure 11B). The EPR spectrum • of YD is distorted at 0.64 mW due to rapid passage effect (the spin-lattice -1 relaxation rate (T1 ) is too slow as compared to the modulation frequency • used) (149). The higher yield of the 20 G radical (“uncoupled” YZ ) in the S1-state compared to S0-state reflects the presence of the S = 0 ground state of the CaMn4-cluster in the S1-state, which will not split the radical in the magnetic interaction, instead resulting in the pure radical spectrum (143), (Paper II).

Figure 11. The light-induced, fast decaying, 20 G radical (A/circles, B/solid line) in • the S1-state compared to YD (A/triangles, B/dashed line). A) The power saturation behavior at 5 K of the two EPR signals present in B measured at the field position of the dotted line in B. B) EPR spectra of the 20 G radical (under illumination minus • after dark decay – difference spectrum) and YD measured at 5 K, microwave frequency 9.27 GHz, microwave power 0.64 mW, modulation frequency 100 kHz, modulation amplitude 3.2 G.

In reference (137), two components, the g = 2.035 peak and the 26 G component (centered at g-2), were resolved in the visible light-induced Split S1 EPR spectrum. These were correlated to the PSII centers that would give rise to the S2 multiline signal (S = ½) and g = 4.1 signal (S = 5/2), respectively, in the S1S2 transition. • The Split S1 EPR signal, reflecting the intermediate S1YZ , is a more accessible EPR probe of the S1-state (Paper V) than the parallel polarization mode EPR signal previously used (135, 150). The study of all these split EPR sig- • nals (SnYZ ) induced at cryogenic temperature are also useful for learning more about the limitations of YZ oxidation in the respective S-states (see chapter 9.2.

• 8.2. YZ formed via the “NIR pathway”

8.2.1. NIR-sensitivity of CaMn4-cluster

The CaMn4-cluster in the S2- and S3-states is sensitive to light in the near- infrared region. This effect was first observed in the S2-state where near- infrared illumination at 150 K induced a spin conversion from the spin state responsible for the S2 multiline (S = ½) into the state responsible for the g = 4.1 EPR signal (S = 5/2) (151). Upon warming to 170 K, the g = 4.1

34 signal conversion is reversed. If the NIR illumination instead occurs at 65 K, resonance appears at g = 10 and 6, which at increasing temperature in the dark change into the g = 4.1 resonance line due to relaxations of the ligand environment thereby changing the zero-field splitting parameters (152). If intact PSII in the S2-state is illuminated with NIR at cryogenic temperatures, a split EPR signal is induced resembling the Split S1 EPR signal (Paper VIII, (153)) (Figure 9, pink – S2). In the S3-state, near-infrared illumination at 4-50 K resulted in the decrease of the S3-state signals from the CaMn4-cluster (g = 6.7-12 in // mode, g = 10 mode) and formation of a g~5 signal and a broad signal around g = 2 (“Split S3” signal, Figure 9, pink – S3) (131, 154, 155). These two signals were assigned to a S2-state of the CaMn4-cluster, a proton deficient S2-state. The two new NIR-induced signals were related to the previously discovered - g~5 EPR signal that is formed during slow decay of the S3…QA -state over several weeks at 77 K, as well as, the g = 2 split EPR signal induced by visible illumination at <30 K in the 77 K-decayed sample (129). The spin state of the CaMn4-cluster was established to be unusually high, S>7/2, from analysis of the isotropic g-value and simulation of the g~5 EPR signal (156). A third way to observe the g~5 signal, was by illumination of the S1-state sample at alkaline pH at -30 °C, which strengthen the assignment of the S2- state to a proton deficient S2-state (155).

8.2.2. Mn excitation The mechanism behind the near-infrared sensitivity is not clear, but it most likely involves absorption by and excitation of the CaMn4-cluster itself. The NIR-induced difference of the CaMn4-cluster between PSII centers displaying the S2 multiline or the g = 4.1 EPR signal is relevant to understanding the mechanism (151). The S2 multiline has been simulated with a Mn4 spin coupling model, where the spin conversion in the S2-state of S = 1/2 to S = 5/2 (corresponding to the multiline signal and g = 4.1 signal, respectively) could be explained by the change in localization of the MnIII ion (MnIII MnIV MnIV MnIII) (157). Three suggested mechanisms of the NIR sensitivity have been presented in the literature; (1) Mn absorption of the NIR light initiates a redistribution of valence between MnIII and MnIV (intervalence transfer) (151), (2) NIR light triggers a spin state transition in the MnIII from S = 2 to S = 1 (158) or (3) the NIR light induce a d-d transition in a MnIII ion causing the spin change (159). The spin coupling model from reference (157) would tend to support intervalence transfer, resulting in a new location of MnIII in g = 4.1. By contrast, in a di-μ-oxo-bridged MnIIIMnIV complex, near-infrared absorption has been assigned to mixed valence bands (160). Common to all three theories is the involvement of a MnIII ion in the ori- III gin of the NIR sensitivity of the CaMn4-cluster. The preference for Mn stems from the observation of a NIR-band in yet another di-μ-oxo-bridged

35 MnIIIMnIV model complex (161) assigned to spin allowed d-d transition in III Mn between the eg orbitals split by the Jahn-Teller distortion. Even in the more asymmetric environment of the CaMn4-cluster an octahedrally coordi- III nated Mn will be Jahn-Teller distorted due to one electron being in the eg levels. This provides an option for explaining the origin of the Mn excitation by NIR light. Boussac et al published (162) action spectra of all NIR- induced signals in the near-infrared region. The action spectra of all the S2- and S3-state signals were similar, hence indicating the same light absorbing species was involved in both S-states. If the mechanism of NIR absorption III involves a Mn , this ion must remain unchanged in the S2S3 transition. III The presence of a Mn even in the S3-state would have consequences for the nature of the oxidation in the S2S3 transition. If we assume that the S2-state III IV IV IV is in the higher Mn4 oxidation state, the observation of NIR sensitivity in both S2- and S3-states would argue against Mn oxidation in the S2S3 transition and support a ligand-based oxidation. However, in MnIIIMnIV model complexes, MnIII as well as MnIV have op- tical features (charge transfer bands and d-d bands) in the NIR region as well as in the visible range (161). The possibility of d-d transitions in MnIV depend strongly on deviation from octahedral symmetry of the ligand field. An asymmetric, restricted ligand sphere of protein ligands, as can be expected in the CaMn4-cluster, could split the energy levels of degenerate d orbitals in MnIV and open up for d-d transitions in the available energy range here also. If we instead consider the NIR absorbing species to be MnIV, a NIR- 3 2 1 triggered d-d transition (t2g t2g eg ) could for example induce (1) electron III transfer from the neighboring Mn to the now empty low-lying d orbital (t2g) of MnIV (thermodynamically favored, occurs at higher T) or (2) oxidation of the nearby YZ (kinetically favored process occurring at lower T). The greater reorganization required to change the ligand coordination sphere of the two Mn ions in (1) could restrict the intervalence transfer to occur only at higher temperatures (T 65 K, found for the spin conversion in the S2-state). At lower temperatures, electron transfer from YZ would instead fill the empty IV low lying d orbital (t2g) in the excited Mn *, requiring reorganization around only one Mn (observed as the induction of the split EPR signal in both the IV S2- and S3-states). If the formation of Mn * is possible, the NIR sensitivity could as well support Mn-centered oxidation in the S2S3 transition for the following reasons: (a) The observation in the S3-state of oxidation of YZ (4- 50 K) (i.e. mechanism 2) instead of intervalence transfer (i.e. mechanism 1) could be explained by the absence of MnIII in this S-state (Mn-based oxidation would result in MnIV IV IV IV). (b) The lack of NIR-induced EPR signals originating from the lower S-states could also indicate that the MnIV that is formed in the S1S2 transition is the NIR sensitive species. This Mn ion is only in the higher oxidation state (IV) in the S2- and S3-states. Looked at in this light, the NIR-sensitivity is not conclusive in arguing for or against Mn or ligand based oxidation in the S2S3 transition.

36 Whatever the mechanism may be, when Mn is excited at cryogenic tem- n • n-1 perature, an electron is transferred from YZ to Mn yielding YZ and Mn . As a result, the CaMn4-cluster goes back one step in the S-state cycle. We have observed induction of the Split S3 EPR signal with monochromatic light in the range 400-900 nm (Paper IV) and could, from the signal + formation in the presence of pre-reduced QA (closed PSII), exclude P680 as the origin of the Split S3 signal in the full spectral range (Paper VI). We could also observe the induction of the accompanying g~5 signal by visible light illumination, which points to the same origin of these signals independent of wavelength of the inducing light. The simplest explanation for these observations would be to extend the limit of the “NIR pathway” to include illumination with wavelength of 400-900 nm. Nevertheless, another possible explanation, which we can neither exclude nor confirm, is that visible monochromatic light induces fluorescence within the dense EPR-sample and the emitted light would then excite in the near-infrared band.

8.2.3. The split EPR signals induced in the S2- and S3-states The split EPR signals in this category were first discovered to be induced by near-infrared illumination (131, 153) and later confirmed to also be inducible with (pure) visible light (Paper I, IV, VI, VIII). The assumed backward electron transfer behavior is only observed in the S2-and S3-states, when the last preceding event in the S-state cycle is electron transfer (S1S2 transition is pure electron transfer step and S2S3 transition is a proton-first and electron-second step, see chapter 5.2). The split • EPR signal induced in the S3-state is the S2YZ intermediate (S2 = S2 minus + H (or S3 plus electron)), and the S2-state would be an isolation of intermediate state between S2 and S3, a deprotonated S2-state (I5), in the proposed extended S-cycle (40, 55). Reduction of the CaMn4-cluster by YZ at cryogenic temperatures seems to be possible in the high potential S-states (S2 and S3) and where no protons are involved in the step to be “reversed”. In Figure 12, we have followed the S-state oscillation of the NIR-induced signals, both looking at the split signals (Figure 12A) and zooming in on the g = 2 region to observe contribution from radical species (Figure 12B). The S2-state dominates in the samples given 1 and 5 turnover flashes and the “Split S2” signal was observed together with a small g = 2 radical. The “Split S3” EPR signal is clearly observed in S3-state dominated samples (2, (3), 6 turnover flashes). A tyrosine like (width and fine structure) g = 2 radical follows the oscillation of the Split S3 signal. The negative radical visible in • the S1-state (0 fl) was assigned to a temperature effect on YD during the • extended 830 nm illumination (decreased with the level of YD ) and was subtracted from the 1-6 flash spectra resulting in the grey spectra, (Figure 12B). The spectra of the oscillating radicals will be analyzed in a forthcom- ing publication.

Figure 12. S-state oscillation of EPR signals induced by 830 nm illumination at 10 K in PSII samples given 0-6 laser flashes to advance in the S-cycle with start in the S1-state. The grey spectra in B are minus 0 flash spectrum (black). The EPR settings as in Fig. 10 except; in A microwave frequency 16 mW , modulation amplitude 10 G; and in B microwave frequency 1 mW, modulation amplitude 3.2 G.

The Split S3 EPR spectrum (Figure 9, pink – S3) is composed of two high field troughs at 3395 and 3425 G, one low field peak 3205 G and an asymmetric derivative shape resonance centered at 3302 G (Paper I). The Split S3 signal has an unusual stability in the dark at cryogenic temperatures compared to the S0 and S1 split signals. The latter signals decay single exponen- tially (t½ = 3 min at 5 K), while the S3 signal decays with multiphase kinetics when induced by visible illumination (Paper I) and shows no decay at all when induced by NIR (Papers IV, VI). The stability of the EPR signal stems from the lack of recombination partners in the NIR illuminated sample. The partial decay of the visible induced EPR spectrum, on the other hand, is due - to recombination with QA that was formed in the same PSII center in the - ox + + stable charge-pair QA /Cytb559 (Car /Chl ). This was proven in experiments with pre-reduced QA, where the NIR-induced Split S3 EPR signal also started - to decay (hence, recombine with QA present from start) (Paper VI).

38 The NIR-induced “Split S2” EPR signal is similar in shape to the Split S1 EPR signal induced by visible illumination in the S1-state (153), but differ a few Gauss in peak position and microwave power saturation, although both fast relaxing signals (Paper VIII). The two signals probably represent the • same S1YZ intermediate, but formed by either forward or backward electron transfer (see above) in response to the different illuminations, visible and NIR, respectively. The NIR induction increased the stability of the signal in the dark at liquid helium temperature because Mn driven chemistry does not - induce the decay partner QA (Paper VIII and Paper VI). Due to the, com- paratively, smaller 20 G resonance at g~2 in the NIR induced spectrum (Fig- • ure 12A, 1fl) reference (137) suggested the NIR-induced S1YZ spectrum represents only those PSII centers that will advance to the S2 multiline form (S = ½) in the S1S2 transition. It is worth noting that visible illumination also can induce a similar split signal from the S2-state (Paper VIII), which we only dare to claim because of the excellent quality of the 1-flashed samples (no miss on the first flash, Paper V). However, the visible-induced signal might be composed of two • forms, where one is the same as the NIR-induced S1YZ intermediate and the • second may be a S2-state signal (S2YZ ) induced at 10 K (Paper VIII).

9. Metalloradical EPR signals III – Methanol and pH dependence of the OEC 9.1. Methanol

Methanol (CH3OH) closely resembles the substrate, water, and has the potential to probe the environment close to the CaMn4-cluster and also function as a substrate analog.

9.1.1. Room temperature

Small alcohols affect the magnetic properties of the CaMn4-cluster in all S- states. The EPR-signals characteristic for the S1- and S3-states disappears in the presence of methanol (135, 136). The spin S = ½ multiline signal in the S2-state is on the other hand enhanced by alcohols, while the g = 4.1 tends to diminish as the concentration of methanol increases (163-165). In the S0- state, a multiline EPR signal appears only in the presence of methanol (166- 168). The methanol dependence of the EPR signals in the S2- and S0-states were investigated and [MeOH]½, the methanol concentration at which half of the spectral change has occurred, was found to be 0.35% (S2 multiline), 0.2% (S2 g = 4.1) and 0.40 % (S0 multiline) in the presence of DMSO (163). These values will be relevant in the following comparison (Paper II). Steady-state oxygen evolution rates have been reported to be more or less unaffected by methanol (163, 169, 170), while misses upon flash illumina-

39 tion increase with increasing methanol concentration, especially in the S2S3 and S3S0 transitions (169). Methanol also influences the relaxation behavior of the EPR signals, for example the microwave power needed for half saturation (P½) of the S2- multiline was found to be 85 mW and 27 mW in the presence and absence of methanol, respectively. From an ESEEM study using 2H-labeled alcohols evidence suggests that methanol directly ligate to the CaMn4-cluster to give 2 the observed effect in the S2-state (170). Findings in a recent H ESEEM study indicate that water is not displaced by methanol in the S0- and S2-states and that instead they observed methanol or water deprotonate upon MeOH- binding (171). Methanol and water were suggested to non-competitively bind to the same Mn with the lowest oxidation state (highest spin state) in both the S0- and S2-states.

9.1.2. Cryogenic temperatures The interesting S-state dependent effects of methanol on the Mn-derived EPR signals described above prompted us to investigate the methanol dependence of the metalloradical EPR signals induced in the S0-, S1-, and S3- states of the CaMn4-cluster (Paper II). • The Split S1 EPR signal (S1YZ ) diminished rapidly with increasing concentration of methanol and half of the amplitude was removed by the presence of 0.12% methanol (Table 1, Paper II). This was no surprise since the S1-state parallel mode signals also vanish in the presence of methanol (135). Both the Split S1 EPR signal (143) and the parallel mode S1-state EPR signal have been simulated to originate from the low-lying excited state S = 1 (separated from the S = 0 ground state by 2.5 K) (135). Methanol has been suggested to increase the Mn–Mn exchange coupling (J, from -0.87 to ~5 cm-1) thereby depopulating the excited spin state (135). In Paper II, we use the same argument of increased Mn–Mn coupling in the presence of methanol to explain the loss of the Split S1 signal that also originates from the S = 1 spin state. • The Split S3 EPR signal (S2YZ ) also vanished with increasing methanol concentration, but with the higher [MeOH]½ of 0.57% (Paper II). Although the Split S3 signal was induced by visible light in Paper II, the pathway of formation of this split signal is different (the “NIR-pathway”) from the + P680 driven formation of the Split S1 signal (Paper VI). The situation for the analysis of methanol dependence of this signal is therefore different. The S2-state is created during the induction of the split signal, while the CaMn4- cluster in S1-state for example remains unchanged during the split signal induction. Methanol can possibly act on three levels to reduce the Split S3 signal intensity: (1) remove the NIR sensitivity by increasing exchange coupling in the starting S3-state, (2) inhibit the backwards electron transfer from YZ, or (3) change the magnetic properties of the S2-state formed.

Figure 13. The Split S0 EPR signal induced in the presence (black) and absence (red) of 5% methanol. EPR conditions: microwave power of 25 mW, microwave frequency of 9.47 GHz, modulation amplitude of 10 G, and temperature of 5 K.

A new broader and more intense form of the Split S0 EPR signal is induced in the presence of methanol, with [MeOH]½ of 0.54% (Paper II) (Figure 13). In methanol-containing PSII samples, the S0 multiline is visible in the dark. The light-induced formation of the Split S0 signal at 5 K occurred with a simultaneous decrease in the S0 multiline spectrum. This suggests that the • same S0-state centers giving rise to the multiline signal can interact with YZ to yield the Split S0 signal. Hence both the S0 multiline and the Split S0 signal originate from the S = ½ ground state of the CaMn4-cluster. An increased Mn–Mn exchange coupling (J) in the presence of methanol would depopu- late the higher spin states, thus increasing the population of the ground state (134). This increased ground state population could explain the more intense resonance of the ground state signal, Split S0, in the presence of methanol, observed in Paper II (Figure 13, black versus red). The broadening of the Split S0 EPR signal in the presence of methanol also suggests a stronger • interaction between YZ and the CaMn4-cluster.

9.1.3. Methanol binding site The methanol dependence of the Mn-related EPR signals place methanol close to the CaMn4-cluster. Methanol resemble structurally substrate water, which raises the possibility that methanol could compete with water for the same sites. Questions arise about the nature of methanol binding: How many sites does methanol bind to? Is it the same site(s) for methanol binding in all S-states? Is it substrate water binding site(s)? In a recent ESEEM study on the S0- and S2- states, only one site for methanol binding was observed, and the authors also concluded that the

41 methanol molecule and the water molecule were binding to separate sites on the same Mn-ion in the S2-state (171). The curves displaying the formation or decay of split EPR signals in the presence of methanol (one [MeOH]½) (Paper II) were also indicative of only one effective binding site. The different [MeOH]½ found in Table 1 and in (163) can be used to di- vide the methanol dependence in two distinct S-state groups: the very methanol sensitive S1- and S2-states ([MeOH]½ = 0.22% ±0.13), and the less sensitive S3- and S0-states ([MeOH]½ = 0.50% ±0.10). During the S1S2 transition, only Mn oxidation occurs, and the CaMn4-cluster is not believed to alter its structure. This seems to be reflected in the similar [MeOH]½ values for both S1- and S2-states. The larger structural changes of the cluster in the S0S1 and S2S3 transitions (43) result in changed accessibility of the methanol binding site. To summarize, the methanol binding site is more accessible/ has a higher affinity in the S1- (0.12%) and S2-states (163) and is decreasingly accessible/ lower affinity in the S0- (0.54%) and S3-states (0.57%). This could indicate separate methanol binding sites in the different S-states, or S-state dependent modifications of the same binding site. It is easiest to assume the same binding site in all S-states considering the pathways for methanol through the protein to approach the CaMn4-cluster. Based on the static PSII structure computer-assisted modeling suggests that only two Mn-ions (3Mn and 4Mn) are accessible to a methanol-sized probe, and considering spectroscopic literature data 3Mn was favored (64). If it is the same methanol site in all S-states, the environment of this Mn must change during the S-state cycle. For example, water molecules can obstruct the site or compete for binding. This could make the methanol binding site less accessible. The higher [MeOH]½ found in the S3- and S0-states might reflect this phenomenon. This adds support to the hypothesis that water binding occurs in two steps, one water in the S4S0 transition and a second water in the S2S3 transition (50, 74). The newly bound water molecule could then reduce the accessibility of the site to methanol, while the presence of methanol in the site could complicate the water binding. The latter might be an explanation for the observed increase of misses in the S2S3 and S3S0 transitions in the presence of methanol in flash oxygen measurements (169). Methanol might not compete directly for the substrate site, but it seems to interfere with the substrate binding.

9.2 pH Investigating the pH dependence of a phenomenon can highlight the importance of protons for the reaction and identify pKa-values of titratable groups involved. The pKa is the negative logarithm of the acid dissociation constant + - Ka, i.e. the equilibrium constant of HA H + A . Amino acids that have ionizable functional groups in the side chain will be titratable in the protein. The pKa-values of the ionizable side chains in the protein can deviate con-

42 siderably from the value measured in aqueous solution. Electrostatic interactions in the folded protein can shift pKa-values up to several pH-units. The redox-active amino acid YZ, involved in the metalloradical intermediates studied in this thesis, is such a titratable amino acid. Depending on the - surrounding pH it has the theoretical reduced forms Tyr-OH, Tyr-O (pKa = + • 10) and the oxidized forms Tyr-OH , Tyr-O (pKa = -2). However, favorable oxidation of tyrosine is coupled to deprotonation (PCET) resulting in the neutral radical. To fully understand the enzymatic mechanism of PSII, the deprotonation–protonation mechanism of YZ-OH during oxidation–reduction is of key importance. The very fast oxidation–reduction cycle in active PSII at physiological temperature aggravates the difficulty in such mechanistic studies. Here, the pH dependence of the split EPR signal formation (YZ oxidation at cryogenic temperature) can be helpful in resolving the deprotonation– protonation of YZ in the intact PSII even in the individual S-states. We should however keep in mind the very low temperature used during split signal induction. At a few degrees Kelvin, proton and protein movements are restricted and there is a frozen-in heterogeneity in PSII centers. To interpret the cryogenic data, it is necessary to first consider what is known about the pH-dependence of electron transfer to and from YZ at room temperature.

9.2.1. Room temperature • The long lived YZ in inactive Mn-depleted PSII have been extensively investigated, but these results will not be reviewed here since the environment around the tyrosine is substantially different without the CaMn4-cluster. In- stead we will focus on the pH-dependence of (i) the re-reduction kinetics of + P680 by YZ and (ii) the reduction of YZ by the CaMn4-cluster in the different S-states in active PSII. + (i) In active PSII, as mentioned before, oxidation of YZ by P680 is multiphasic, S-state dependent and occurs with both ns- and μs-kinetics (23). The pH dependence of P680+ reduction in active PSII has been investigated several times (98, 172-174). The fast ns-kinetics component, which dominates in the S0- and S1-states, decrease both in amplitude and speed in the acidic region and is virtually unaffected in the alkaline region (98, 172). The ns- kinetics are suggested to reflect a proton shifting in a well-tuned hydrogen bridge between YZ-OH and a nearby base (93, 97, 98), identified as D1- His190 (HisZ) (4, 24). Protonation of this base, would prevent the formation of the well-tuned hydrogen bond and thereby the fast ns-kinetics. A pKa ~4.5-5.3 was reported for the decrease of the fast ns-kinetics in active PSII (172-175), which most likely reflects the protonation of the imidazole N of HisZ (98, 175). The pH dependence of the slower ns-kinetics, dominating YZ oxidation in the higher S-states and reflecting a dielectric relaxation response near YZ (23,

43 176), has not been directly addressed. The much slower μs-component in oxygen-evolving PSII (23) was shown to oscillate in size with the S-states (94, 177). This μs-component shows a marked H/D isotope exchange effect, especially in the higher S-states, and is suggested to be coupled to a large scale proton relaxation involving proton movement in a hydrogen bond network (97, 177). By comparing the extent of turnover-misses versus pH in both D2O and H2O, the deviation in misses, ((D)/(H) < 1) at alkaline pH led to the conclusion that the hydrogen bond network was deprotonated and that the “slow μs-kinetics” were set out of play at pH >7 (98). The pH dependence of the multiphasic P680+ reduction kinetics analyzed in all the separate S-state has not been published. (ii) Let us consider reduction of YZ by the CaMn4-cluster in the different S-states. The information from the S-state transitions can be translated into whether YZ is operational. YZ oxidation is a prerequisite of S-state transitions, even though the reverse is not true. The pH dependence of the S-state transitions have been investigated with EPR (spinach) (178) and FTIR (T. elongatus) (179). The conclusions in these studies were similar and will be treated in one overview.

S0S1 This transition was inhibited at acidic pH (pKa ~4.7), but above this the S1-state was efficiently formed independent of pH (up to pH < 9.5).

S1S2 This transition was open in the entire pH interval investigated (pH 4-9). This is a critical point and means that YZ has the potential to function in this pH range. However, these studies do not distinguish between fast or slow kinetics.

S2S3 The transition is inhibited at acidic pH (pKa ~3.8±0.2) and shows a slight decrease at high pH (pKa ~9.4).

S3S0 This transition, that includes O2 formation and binding of water, is inhibited both at low (pKa ~4.5/4.2) and high pH (pKa ~8.0/10.2). The inhibition at high pH was assigned to a • decrease in the redox potential of the YZ /YZ couple. The difference between spinach and T. elongatus was explained by the generally lower redox potentials assumed for the S-states in T. elongatus.

The acidic pKas found in S0S1, S2S3 and S3S0 were likely to be the result of the inhibition of the proton release that is necessary for these transitions. Protonation of amino acid residues involved in the proton exit channels is a possible cause. Another process that can occur at acidic pH is the removal of Ca2+ from the cluster (180). In (178), the acidic release of Ca2+ as the cause of the acidic pKs was excluded due to the use of short incubation time at moderately low pH (4-5) in presence of 5 mM CaCl2.

44 9.2.2. Cryogenic temperatures • We investigated the pH dependence of YZ formation at cryogenic temperatures (split signal formation) in the S1-state (Paper III), S0-state (Paper VII) and S3-state (preliminary data in Figure 14). The Split S1 EPR signal was also employed to study the pH-dependent competition between YZ and YD oxidation at cryogenic temperatures in the S1-state (Figure 15B) (Paper III). Oxidation of YZ-OH is coupled to deprotonation, which is limited by the low temperature (5 K). The PCET reaction at 5 K will take place only in PSII with YZ frozen in an H-tunneling ready configuration that does not require extensive proton movements. The fairly short (2.66-2.78 Å in the static PSII structure (4, 5, 24)) well-set H-bond between YZ-O-H….HisZ seems to support proton tunneling at 5 K and is assumed to be the prerequisite for split signal induction. A certain percentage of the PSII centers will be frozen in this favorable configuration, which allows YZ oxidation at 5 K (observed as the split EPR signal). Titration of this important H-bond may have been observed in our pH-studies of the split signals (Paper III, VII).

Split S0 Split S1 Split S3

Figure 14. The pH dependence of the three split EPR signals induced by illumination (Split S0 and Split S1 – visible light, Split S3 – 830 nm light) at 5 K. The red lines are fitted to the one (S1) or two pKs (S0 and S3). (Paper VII, III and unpublished data)

Acidic region: In this region we found the same acidic onset (apparent pKa ~5) of cryogenic YZ oxidation (split signal formation) independent of S-state (Figure 14), and we correlated this to a titration event disrupting the essential H-bond to HisZ prior to freezing (Paper III). The simplest explanation is direct protonation of N in the imidazole ring of HisZ that is the hydrogen bond acceptor (paper III). The same well-set H-bond was described to be a prerequisite of the fast ns-kinetics observed for the reduction of P680+, which also decreased with a similar pKa as we found (~4.5-5.3 (172-175) versus ~5, respectively) for the decrease of the split signal formation. This is not the first observation of a correlation between fast electron donation to P680+ (ns-kinetics) and cryogenic oxidation of a hydrogen bonded tyrosine.

45 • It has also been observed in the formation of YD within t½ = 190 ns at pH 8.5 (181) and by illumination at 15 K at pH 8.5 (182). We found the acidic decrease of YZ oxidation (cryogenic and fast ns component) to have the pKa ~5, which deviates approximately one pH unit from the imidazole pKa = 6.04 in solution. This is not strange considering the structural and electrostatic influence of the neighboring CaMn4-cluster that can well render the pKa of the imidazole more acidic. For example, in the absence of the CaMn4-cluster a pKa ~7-7.5 has been observed for fast elec- + tron transfer to P680 (t½ ~1 μs) (174, 183), which is thought to reflect pro- 2+ 2+ tonation of a base (HisZ (91)). However, binding of Ca or Mg to the Mn- depleted PSII was observed to lower the pKa to ~6.6 for the fast kinetic phase (184). However, at room temperature the S1S2 transition is open also at low pH, which reason that YZ oxidation is also functional under these conditions. The explanation for these seemingly contradicting data from cryogenic temperature and room temperature experiments can be found in the multiphasic kinetics of P680+ reduction. The μs-kinetics that is thought to require deprotonation of YZ in a H-bond network remained unaffected at low pH (98) and is functional at room temperature. In contrast, these extensive proton movements can be expected to freeze out at 5 K thereby preventing the formation • of YZ at pH < pKHis-Z (Paper III).

46 A – Split S0 B –Split S1 C –Split S3

Figure 15. The pH dependence of EPR signals between pH 6-9. A – The decrease of the Split S0 amplitude (black) and fast decaying radical (white) and increase of the broad g-2 signal (blue) (Paper VII). The dashed lines between white and black circles represent, measured in the same PSII sample. B – The amplitude of the Split S1 signal in PSII with YD-oxidized (black) and YD-reduced (white) compared to the • increase of the light induced stable YD (gold) (Paper III). C – The amplitude of the NIR induced Split S3 EPR signal (black) decrease together with the NIR-induced radical (white) (unpublished data). Present in the dark before illumination is the alkaline-induced split signal (green) (185).

Alkaline region: The three split EPR signals studied behaved differently in the high pH region (Figure 14) and will therefore be treated separately. In Paper VII, we observed that the formation of the Split S0 EPR signal decreases at higher pH (black circles, Figure 15A). The S0-state sample also allowed us to follow the induction of a fast decaying radical (white circles, Figure 15A) that we, from its fast relaxation (P½ = 0.3 mW at 5 K), signal • shape (20 G) and g-value (2.0046), assigned to uncoupled YZ in 5% of PSII (Paper VII). This radical followed the exact pH dependence of the Split S0 signal in the low pH region (4.0< pH >6.5), supporting our assignment of pKa ~5 to a break in the important H-bond for YZ oxidation. However, on the high pH side, the two EPR signals initially decreased together (7.0< pH >7.5), but at pH > 7.5 the radical (white circles) stayed high while the Split S0 signal (black circles) continued to decrease (clarified by the dashed lines, Figure 15A). A plausible explanation of the initial decrease of YZ oxidation at pH 7.0- 7.5 (radical and split signal together) is deprotonation of a stabilizing hydrogen bond network around YZ and HisZ, involving amino acid residues (e. g. D1Gln165, D1Asn296, D1Glu189, D1Asp170 around YZ (Figure 16, left) and D1Asn298, CP43-Gly402 around HisZ (Figure 16, right)) and water(s).

47 Depleting this hydrogen bond network of protons could result in destabiliza- tion of the YZ-OH…N-HisZ bond and extend its flexibility to form other H- bonding interactions, which in turn would increase the probability of freezing-in the H-bond in a less optimal configuration in an increased number of PSII centers (Figure 17, left panels). This will not allow proton coupled cryogenic oxidation of YZ to the same extent at pH > 7. This interpretation correlates with the observed deviation in misses in S-state transitions between D2O and H2O PSII samples at pH > 7.0 (98) (see detailed comparison in Paper VII), which was explained by deprotonation of several weak acids, that form the H-bond network around YZ.

Figure 16. In the left panel, a suggestion of the stabilizing hydrogen bond network around YZ to involve the polar or charged amino acids, D1Gln165, D1Asn296, D1Asp170, D1Glu189, and water molecule(s) in the space between YZ and OEC (shaded area). In the right panel, a possible hydrogen network around HisZ with H- bonds (turquoise dashed lines) between YZ…D1His190…D1Asn298…CP43- Gly402. (PDB file 3BZ1)

48 4.7 < pH > 7.0 pH < 7.7 H Y l Z O H N- Mn Mn O-H N D1His190

B-H

pH > 7.0 pH > 7.7

YZ O Mn Mn H N- O-H N D1His190

B-

Figure 17. Mechanistic explanations for the decrease of the Split S0 EPR signal. Upper left – At pH between 4.7 and 7.0 YZ is easy to oxidize at cryogenic temperature due to the well-tuned H-bond (YZ…HisZ) and the well set H-bonding network around YZ-O (YZ…HB). Lower left – At pH above 7.0 the H-bonding network starts to deprotonate (light grey area) and YZ oxidation is less optimal. Upper right – At pH < 7.7 is there a μ-hydroxo-bridge in the CaMn4-cluster (protonated state). Lower right – At pH >7.7 the CaMn4-cluster itself is deprotonated (deprotonation to a μ-oxo-bridge). We suggest that this leads to changes the magnetic properties of the cluster, resulting in a new type of interaction signal, broad g-2.

At high pH (pKa ~8), a new EPR signal (blue circles, Figure 15A) is induced by visible illumination at 5 K, at the same time that the amplitude of the Split S0 peaks decrease. The new signal looks like a broadening of the g~2 component (or a very narrow splitting), hence the name “broad g-2”. The spectrum of the broad g-2 signal is hard to saturate with microwave power and decays in the dark with t½ = 3 min (Figure 18). The fast relaxation property of the broad g-2 signal points to the involvement of transition metals - and its decay rate points to charge recombination with QA in the dark, coherent to the other split EPR signals investigated (Paper I). We therefore propose in line with the other split signals, the broad g-2 signal arise from • YZ being in magnetic interaction with a deprotonated CaMn4-cluster (Figure 17, right panels). The amplitude of the S0 multiline signal decreased at alkaline pH with a pKa ~8.0 (186), similar to the induction of the new broad g-2 signal (Figure 15A, blue circles), even though the S0S1 transition is open at high pH (178). There is most likely a protonated μ-oxo-bridge present in the S0-state (39, 43) and deprotonation of the bridge would change the distance between the Mn-ions (43), which can change the magnetic coupling within the cluster. This provides a plausible explanation for the lost S0 multiline amplitude and the changed shape of the Split S0 signal at high pH. Both processes are reversible by changing the pH back to pH~6 (re-protonation of

49 the bridge) ((186), Paper VII). Curiously enough, the similarity in signal shape between the broad g-2 and the split signal formed in Ca2+-depleted PSII by illumination at 243 K (126) (Figure 9, white – S1) is striking.

Figure 18. A broad g-2 EPR signal induced by visible illumination at 10 K in a S0- state sample at pH 8. The decaying signal, light minus after-dark-decay spectrum is presented, measured at microwave power 81 mW (blue), 20 mW (red) and 5 mW (black). EPR conditions: Microwave frequency 9.27 GHz, field modulation frequency 100 kHz, field modulation amplitude 10 G, temperature 10 K.

The formation of Split S1 EPR signal was pH independent above pH 5 (pKa ~4.7) if YD was oxidized from start (Paper III). The situation was however different in the presence of reduced YD and in Figure 15B shows the steep drop in the Split S1 signal with increasing pH (pKa ~7.9) (white circles) and • the EPR spectrum of the light-induced stable YD radical appeared with a similar pKa ~8 (gold circles) (Paper III). These pKas are similar to literature pKa ~7.6 for YD oxidation at cryogenic temperature (182). Reduced YD seems therefore to be the preferred donor to P680+ when it is available at alkaline pH. For detailed discussion see Paper III. The analysis of the pH dependence of the NIR-induced Split S3 EPR signal in the high pH region (Figure 15C, black circles) was complicated by the formation of the alkaline-induced split signal already present in the dark • at room temperature. This alkaline-induced split signal was assigned to YZ • formed by decreasing the potential of the YZ /YZ redox couple at alkaline • pH, thereby shifting the S3YZ – S2 YZ equilibrium towards a proton deficient • • S2 YZ (185). It means that YZ is already present before split signal induction and hence cannot be formed at 5 K. This is most likely the reason for the apparent decrease of the Split S3 signal formation (black circles, Figure 15C) with a similar pKa to the increase of the alkaline-induced split signal (green circles, Figure 15C). From the preliminary data in Figure 15 C, the NIR-

50 induced tyrosine-like radical (Figure 12, 2-fl right panel) decreased at alkaline pH in concert with the Split S3 EPR (white and black circles, Figure 15C). This strengthens the assignment of the alkaline decrease to the un- availability of reduced YZ for light-induced oxidation at 5 K.

10. Conclusions and future perspectives

10.1. Concerning YZ oxidation

The fact that YZ is oxidizable at cryogenic temperatures in intact PSII was not obvious to the field ten years ago, at the time of the first observations of a light-induced split signal at cryogenic temperture (129). A prerequisite to • cryogenic induction of YZ in intact PSII, is the very important H-bond to HisZ that must be well-set and supported in a hydrogen-bond network in order to be functional in deprotonating YZ at 5 K. The high yields of the split signals (~50 % of PSII) indicate a well-ordered donor side at physiological condition, with YZ positioned for optimal performance in the intact functional PSII. The previously suggested disorder around YZ (81) compared to YD (82) stems from studies in Ca/Mn depleted PSII (187). These studies are fundamentally very interesting, but with the catalytic site removed from PSII, they do not report on the actual water splitting mechanism. The organization of the functional donor side involves an intricate balance between the amino acid residues, water molecules and the inorganic cofactors that cannot be observed in depleted inactive PSII.

10.2. Concerning the metalloradical EPR signals The “double nature” of the metalloradical EPR signals allows us to investi- • gate both YZ and the CaMn4-cluster. In general, any effect on the oxidation of YZ changes the amplitude of the split signal, for example the decrease of Split S1, Split S0 and Split S3 at acidic pH. Only when YZ oxidation at cryogenic temperatures is functional can the split signals be used to probe changes on the CaMn4-cluster, by for example observing the change in the shape of the Split S0 signal in the presence of methanol (the NIR-induced signals are an exception). All of the S-state dependent split EPR signals can however be very useful probes for the quantification of S-states in mixed PSII samples. Even though the split EPR signal is not formed in all PSII centers, the propor- tions are well correlated to the S-state population (Paper V). These light-induced water oxidizing intermediates isolated at cryogenic temperatures from intact PSII have the potential to be explored with more techniques. This could yield additional informative spectroscopic signatures that can deepen our understanding of the still challenging mechanism of water oxidation.

51 Acknowledgement

I would like to thank the following people that made these years possible:

Stenbjörn Styring, my supervisor and mentor, thank you for introducing me to the amazing photosystem II, EPR and to life as a scientist from the first inspiring lecture in Lund all the way to this thesis in Uppsala. Thank you for both challenging and believing in me. Fikret Mamedov, my co-supervisor and man in the lab, thank you for al- ways having time for my questions and discussions and teaching me spectroscopy. Felix Ho, a big THANK YOU. I have learnt a lot from you during the years as your desk neighbour and collaborator. You are a great friend and thanks again for the last desperate days… hours. Anders Thapper, thank you for all the broad scientific discussions and for yours and Patricia’s friendship. Philipp Kurz, thank you for making the days at fotomol brighter, showing us Sweden, and for all movie Sundays at Rackarberget. Johannes Sjöholm, Guangye Han and Ji-Hu Su thank you for good col- laboration and shared interest in the split EPR signals. Åsa Söderberg and Tanai Cardona, thank you for the good atmosphere in our office and many nice chats. Gustav Berggren for showing interest in my NIR-induced signals and helping me understand. Gunilla Hjort, tack för gott sällskap på morgonen och för allt ditt arbete med spenaten. Gabor Bernat and Yashar Feyziyev, thank you for thorough education on low temperature EPR measurements when I was a new student in Lund. To all other past and present members of our group: Fredrik Andersson, Clyde Cady, Guiying Chen, Ravi Danielsson, Ping Huang, Joakim Högblom, Jordanos Kiflemariam, Ann Magnuson, Susan F. Morvaridi, Erik Nagel, Nizamuddin Shaikh, Denys Shevchenko, Wei Shi, Hannes Uchtenhagen, Beatriz Villaroel, Michael Wild and Chunxi Zhang. Thank you for all the intense discussions during our meetings and good times. Thank you to all past and present members at the department of Photo- chemistry and Molecular Science for the great atmosphere you create. Thanks to the people at the Biochemistry department in Lund in year 2003. I had a great time with you.

52 Tack till mina föräldrar och bror för att ni gett mig den trygget det är att alltid vara älskad. Ni gör mig stark. Ni ställer alltid upp, även nu med pass- ning av Freja. Freja min lilla flicka, tack för din soliga uppenbarelse som fyller mig med sådan glädje och energi genom att bara finnas till. Mamma ska bli mer när- varande nu. Tillsist, Sven min älskling, jag är så lycklig att ha dig vid min sida. Med din kärlek kan jag klara allt. TACK!

53 54 Svensk sammanfattning

Fotosyntesen är i växter, alger och cyanobakterier den viktiga enzymatiska processen i naturen som en gång för alla förändrade vår jord. När cyanobakterierna började uppträda för cirka 3 miljarder år sedan möjliggjorde de ut- vecklingen av alla de livsformer som andas syre, däribland människan, genom att totalt förändra jordatmosfärens sammansättning. Från att bestått av framförallt koldioxid och vattenånga ändrades atmosfärens sammansättning till att idag innehålla 21 % syre. Detta gjordes genom fotosyntesen, den process där solens energi binds upp i energirika molekyler, enligt den generella formeln:

6 H2O + 6 CO2 + solljus C6H12O6 (socker) + 6 O2

Cyanobakterierna frisatte O2 (syre), som en biprodukt i processen, genom sin användning av solljuset som energikälla. Fotosystem II som i en ljusdriven reaktion spjälkar vatten och släpper ut syrgas är det första enzymet i den fotosyntetiska reaktionskedjan. Fotosystem II och vattenspjälkningen är i fokus för denna avhandling. Jag har studerat hur mekanismen för vatten- spjälkningen går till på molekylär nivå. Varför är då detta så intressant att förstå? Det är naturligtvis alltid intressant att förstå hur naturen fungerar på ett grundforskningsplan, men jag tror också att fotosystem II än en gång har chansen att förändra livet på jorden. Jag tänker på vårt ständigt ökande be- hov av energi som står i konflikt med att olja och kol, de vanligaste energi- bärarna idag, är ändliga resurser. Dessutom vid förbränningen av kol och olja frigörs den koldioxid som en gång bands i kolföreningarna i växter och bakterier. Koldioxid är en effektiv växthusgas som bidrar till den globala uppvärmningen. Detta tillsammans med den begränsade tillgången av fossila bränslen är två goda skäl till att minska vårt beroende av fossila energikällor. Vad vi behöver är en ny energibärare till en gammal energikälla (det är sol- energi som är lagrad i det fossila bränslet via fotosyntesen). Tänk om vi skulle kunna binda solens energi i en energirik molekyl i ett eget fotosystem utan att behöva gå omvägen via växternas fotosyntes. En energibärare som diskuteras idag och som har en potentiell framtid är vätgas. Tänk om vi kan koppla vätgasproduktion till en ljusdriven process där solens energi fångas till att spjälka vatten till sina beståndsdelar. Vattnets elektroner och protoner skulle sedan kombineras igen till lättillgänglig energi i form av vätgas. För

55 att kunna tillverka en sådan katalysator kan det vara en genväg att studera de mekanistiska principer som gäller hos den biologiska katalysatorn för vatten- spjälkning. Det är här studier av fotosystem II kommer in i bilden. Växter, alger och cyanobakterier har redan en katalysator för denna process, fotosystem II. Det är i stora drag känt hur detta enzym fungerar, men när det kommer till de molekylära detaljerna kring hur elektroner och protoner frigörs från vatten för driva andra reaktioner i cellen så är bilden fortfarande oklar. Det är på denna sista pusselbit som jag har forskat. Följande händer i fotosystem II efter det att solljuset absorberas av kloro- fyllerna, de gröna molekylerna i växterna, som sitter i antennprotein. En klorofyllmolekyl i hjärtat av fotosystem II (P680) uppnår en högre energini- vå, exciteras (P680*) och kan avge en elektron till en väntande acceptormo- lekyl, som skickar elektronen vidare (elektronen kommer efter en lång väg i cellen ingå i energirika föreningar). Det har bildats ett elektron-hål hos klo- rofyllmolekylen (P680+) som drar till sig en ny elektron från en aminosyra- • rest i närheten, Tyrosin-Z (YZ). En instabil radikal har bildats, YZ . Radikalen tar en elektron från/oxiderar den angränsande CaMn4-komplexet som utgör det katalytiska sätet i enzymet. Vatten som har bundit in här är den slutgilti- ga elektronkällan. För att avge en molekyl syrgas (O2), så behöver 2 H2O oxideras i en fyrastegsprocess, dvs. fyra gånger måste P680 exciteras och fyra elektroner lämnar CaMn4-komplexet. Detta sker i en cyklisk process (S- tillstånds-cykeln), där oxidationstillståndet på mangan ändras, dvs. fyra posi- tiva laddningar ackumuleras på CaMn4-komplexet (S1, S2, S3, S4). I det sista steget (S4) frisläpps O2 och Mn-jonerna återgår till sin ursprungliga laddning (S0) med ny inbindning av vatten. Det verkar kanske som om vi vet mycket om denna komplexa process, men det återstår många avgörande detaljer att utforska innan vi kommer ha en full förståelse för hur processen går till. Vi har karaktäriserat nya mellanliggande steg som ökar detaljrikedomen i bilden av mekanismen. Med elektron-paramagnetisk-resonans spektroskopi (EPR) kan man se oparade elektroner, spin (1 elektron = 1 spin) som finns i • fria radikaler som YZ och i övergångsmetaller som mangan. CaMn4- komplexet ger karaktäristika signaler beroende de ingående jonernas oxida- • tionstillstånd (och spin-tillstånd). För att kunna studera YZ radikalen och dess roll i vatten oxidationen så har vi kylt ner vårt prov till 5 K. Bara fem grader från den absoluta nollpunkten har rörelser i proteinet och längre för- flyttningar av protoner frusit ut. Elektroner kan fortfarande röra sig, men • elektrontransport från CaMn4-komplexet är blockerad. Vi kan fånga YZ efter belysning vid 5 K. Vi har nu möjlighet att studera radikalen och dess omgiv- ning med hjälp av de ”spin-spin interaktions-EPR-signaler” (kallade split • signals i avhandlingen) som uppstår mellan den oparade elektronen på YZ • och de oparade elektroner som finns i CaMn4-komplexet. Vi kan fånga YZ i de olika S-tillstånden och får då karaktäristiska EPR-signaler för varje S- tillstånd (Figur 1). Vi har gjort mekanistiska studier av dessa signaler och • lärt oss om vätebindningsnätverket runt YZ som verkar vara väl organiserat

56 • och inte alls ostrukturerat som tidigare antagits. Dessa tidigare studier av YZ har skett i fotosystem II där CaMn4-komplexet tagits bort. Men att från detta • system dra slutsatser om YZ och dess nära samarbete med CaMn4-komplexet är inte självklart. Därför tror jag att fortsatta studier av dessa nya EPR signaler i intakt, i syrgasutveckling, aktivt fotosystem II kan ge fler svar om me- • kanismen för vattenspjälkning och framför allt vilken roll som YZ spelar.

Figur 1: I bilden visas S-tillstånds-cykeln och mellan vilka S-tillstånd man tror att elektroner och protoner lämnar CaMn4-komplexet. I respektive tårtbit finns spektra med de karaktäristiska EPR signalerna från de olika S- tillstånden som har studerats i avhandlingen.

57 References

1. IEA (2008) World Energy Outlook 2008, OECD/International Energy Agency, Paris. 2. Lewis, N. S., and Nocera, D. G. (2006) Powering the planet: Chemical challenges in solar energy utilization, Proc. Natl. Acad. Sci. U.S.A. 103, 15729-15735. 3. Hill, R., and Bendall, F. (1960) Crystallization of a photosynthetic reductase from a green plant, Nature 187, 417. 4. Loll, B., Kern, J., Saenger, W., Zouni, A., and Biesiadka, J. (2005) To- wards complete cofactor arrangement in the 3.0 Å resolution structure of photosystem II, Nature 438, 1040-1044. 5. Guskov, A., Kern, J., Gabdulkhakov, A., Broser, M., Zouni, A., and Saenger, W. (2009) Cyanobacterial photosystem II at 2.9-Å resolution and the role of quinones, lipids, channels and chloride, Nature structural & molecular biology 16, 334-342. 6. Prokhorenko, V. I., and Holzwarth, A. R. (2000) Primary processes and structure of the photosystem II reaction center: A photon echo study, Jour- nal of Physical Chemistry B 104, 11563-11578. 7. Rigby, S. E. J., Nugent, J. H. A., and Omalley, P. J. (1994) ENDOR and special triple-resonance studies of chlorophyll cation radicals in Photosys- tem-2, Biochemistry 33, 10043-10050. 8. Szczepaniak, M., Sugiura, M., and Holzwarth, A. R. (2008) The role of TyrD in the electron transfer kinetics in Photosystem II, Biochim. Biophys. Acta 1777, 1510-1517. 9. Ishikita, H., Loll, B., Biesiadka, J., Saenger, W., and Knapp, E. W. (2005) Redox potentials of chlorophylls in the photosystem II reaction center, Bio- chemistry 44, 4118-4124. 10. Jursinic, P., and Govindjee. (1977) Temperature-dependence of delayed light-emission in 6 to 340 microsecond range after a single flash in chloroplasts, Photochemistry and Photobiology 26, 617-628. 11. Klimov, V. V., Allakhverdiev, S. I., Demeter, S., and Krasnovskii, A. A. (1979) Photo-reduction of pheophytin in the photosystem-2 of chloroplasts depending on the oxidation-reduction potential of the medium, Doklady Akademii Nauk Sssr 249, 227-230. 12. Rappaport, F., Guergova-Kuras, M., Nixon, P. J., Diner, B. A., and Lavergne, J. (2002) Kinetics and pathways of charge recombination in photosystem II, Biochemistry 41, 8518-8527. 13. Grabolle, M., and Dau, H. (2005) Energetics of primary and secondary electron transfer in Photosystem II membrane particles of spinach revisited

58 on basis of recombination-fluorescence measurements, Biochim. Biophys. Acta 1708, 209-218. 14. Gibasiewicz, K., Dobek, A., Breton, J., and Leibl, W. (2001) Modulation of primary radical pair kinetics and energetics in photosystem II by the redox state of the quinone electron acceptor Q(A), Biophys. J. 80, 1617-1630. 15. Eckert, H. J., Wiese, N., Bernarding, J., Eichler, H. J., and Renger, G. - (1988) Analysis of the electron-transfer from Pheo to QA in PSII membrane-fragments from spinach by time resolved 325 nm absorption changes in the picosecond domain, FEBS Lett. 240, 153-158. 16. Krieger, A., Rutherford, A. W., and Johnson, G. N. (1995) On the determination of redox midpoint potential of the primary quinone electron- acceptor, Q(a), in photosystem-II, Biochim. Biophys. Acta 1229, 193-201. 17. Krieger-Liszkay, A., and Rutherford, A. W. (1998) Influence of herbicide binding on the redox potential of the quinone acceptor in photosystem - II. Relevance to photodamage and phytotoxicity, Biochemistry 37, 17339- 17344. 18. Stowell, M. H. B., McPhillips, T. M., Rees, D. C., Soltis, S. M., Abresch, E., and Feher, G. (1997) Light-induced structural changes in photosynthetic reaction center: Implications for mechanism of electron-proton transfer, Science 276, 812-816. 19. Hienerwadel, R., and Berthomieu, C. (1995) Bicarbonate binding to the non-heme iron of photosystem II investigated by Fourier transform infrared difference spectroscopy and C-13-labeled bicarbonate, Biochemistry 34, 16288-16297. 20. Petrouleas, V., and Diner, B. A. (1987) Light-induced oxidation of the acceptor-side Fe(II) of photosystem-II by exogenous quinones acting through the QB binding-site .1. Quinones, kinetics and pH-dependence, Biochim. Biophys. Acta 893, 126-137. 21. Diner, B. A., and Petrouleas, V. (1987) Light-induced oxidation of the acceptor-side Fe(II) of photosystem-II by exogenous quinones acting through the QB binding-site .2. Blockage by inhibitors and their effects on the Fe(III) electron-paramagnetic-res spectra, Biochim. Biophys. Acta 893, 138-148. 22. Berthomieu, C., and Hienerwadel, R. (2001) Iron coordination in photosystem II: Interaction between bicarbonate and the Q(B) pocket studied by Fourier transform infrared spectroscopy, Biochemistry 40, 4044-4052. 23. Brettel, K., Schlodder, E., and Witt, H. T. (1984) 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 O2-evolving complex, Biochim. Biophys. Acta 766, 403-415. 24. Ferreira, K. N., Iverson, T. M., Maghlaoui, K., Barber, J., and Iwata, S. (2004) Architecture of the photosynthetic oxygen-evolving center, Science 303, 1831-1838. 25. Kok, B., Forbush, B., and McGloin, M. (1970) Cooperation of charges in photosynthetic O2 evolution-I. A linear four step mechanism, Photochem. Photobiol. 11, 457-475.

59 26. Hanley, J., Deligiannakis, Y., Pascal, A., Faller, P., and Rutherford, A. W. (1999) Carotenoid oxidation in photosystem II, Biochemistry 38, 8189- 8195. 27. Buser, C. A., Diner, B. A., and Brudvig, G. W. (1992) Photooxidation of cytochrome-b559 in oxygen-evolving photosystem-II, Biochemistry 31, 11449-11459. 28. Kaminskaya, O., Shuvalov, V. A., and Renger, G. (2007) Evidence for a novel quinone-binding site in the photosystem II (PS II) complex that regu- lates the redox potential of cytochrome b559, Biochemistry 46, 1091-1105. 29. Faller, P., Fufezan, C., and Rutherford, A. W. (2005) Side-path electron donors: cytochrome b559, chlorophyll Z and β-carotene, in Photosystem II: The light-driven water:plastoquinone oxidoreductase (Wydrzynski, T., and Satoh, K., Eds.) pp 347-365, Springer, Dordrecht. 30. Bondarava, N., De Pascalis, L., Al-Babili, S., Goussias, C., Golecki, J. R., Beyer, P., Bock, R., and Krieger-Liszkay, A. (2003) Evidence that cytochrome b559 mediates the oxidation of reduced plastoquinone in the dark, Journal of Biological Chemistry 278, 13554-13560. 31. Tracewell, C. A., and Brudvig, G. W. (2008) Multiple Redox-Active Chlorophylls in the Secondary Electron-Transfer Pathways of Oxygen- Evolving Photosystem II, Biochemistry 47, 11559-11572. 32. Ananyev, G. A., Sakiyan, I., Diner, B. A., and Dismukes, G. C. (2002) A functional role for tyrosine-D in assembly of the inorganic core of the water oxidase complex of photosystem II and the kinetics of water oxidation, Biochemistry 41, 974-980. 33. Styring, S., and Rutherford, A. W. (1987) In the oxygen-evolving complex + of photosystem II the S0 state is oxidized to the S1 state by D (Signal- IIslow), Biochemistry 26, 2401-2405. 34. Yano, J., Kern, J., Irrgang, K. D., Latimer, M. J., Bergmann, U., Glatzel, P., Pushkar, Y., Biesiadka, J., Loll, B., Sauer, K., Messinger, J., Zouni, A., and Yachandra, V. K. (2005) X-ray damage to the Mn4Ca complex in single crystals of photosystem II: a case study for metalloprotein crystallography, Proc. Natl. Acad. Sci. U.S.A. 102, 12047-12052. 35. Grabolle, M., Haumann, M., Muller, C., Liebisch, P., and Dau, H. (2006) Rapid loss of structural motifs in the manganese complex of oxygenic photosynthesis by x-ray irradiation at 10-300 K, Journal of Biological Chemis- try 281, 4580-4588. 36. Yano, J., Kern, J., Sauer, K., Latimer, M. J., Pushkar, Y., Biesiadka, J., Loll, B., Saenger, W., Messinger, J., Zouni, A., and Yachandra, V. K. (2006) Where water is oxidized to dioxygen: structure of the photosynthetic Mn4Ca cluster, Science 314, 821-825. 37. Sproviero, E. M., Gascon, J. A., McEvoy, J. P., Brudvig, G. W., and Ba- tista, V. S. (2008) Quantum mechanics/molecular mechanics study of the catalytic cycle of water splitting in photosystem II, J. Am. Chem. Soc. 130, 3428-3442. 38. Siegbahn, P. E. M. (2008) Mechanism and energy diagram for O-O bond formation in the oxygen-evolving complex in photosystem II, Philos. Trans. R. Soc. B-Biol. Sci. 363, 1221-1228.

60 39. Kulik, L. V., Epel, B., Lubitz, W., and Messinger, J. (2007) Electronic structure of the Mn4OxCa cluster in the S0 and S2 states of the oxygen- evolving complex of photosystem II based on pulse 55Mn-ENDOR and EPR Spectroscopy, J. Am. Chem. Soc. 129, 13421-13435. 40. Dau, H., and Haumann, M. (2007) Time-resolved X-ray spectroscopy leads to an extension of the classical S-state cycle model of photosynthetic oxygen evolution, Photosynth. Res. 92, 327-343. 41. Kulik, L. V., Epel, B., Lubitz, W., and Messinger, J. (2005) 55Mn-pulse ENDOR at 34 GHz of the S0 and S2 states of the oxygen-evolving complex in photosystem II, J. Am. Chem. Soc. 127, 2392-2393. 42. Roelofs, T. A., Liang, W. C., Latimer, M. J., Cinco, R. M., Rompel, A., Andrews, J. C., Sauer, K., Yachandra, V. K., and Klein, M. P. (1996) Oxi- dation states of the manganese cluster during the flash-induced S-state cycle of the photosynthetic oxygen-evolving complex, Proc. Natl. Acad. Sci. U.S.A. 93, 3335-3340. 43. Haumann, M., Muller, C., Liebisch, P., Iuzzolino, L., Dittmer, J., Grabolle, M., Neisius, T., Meyer-Klaucke, W., and Dau, H. (2005) Structural and oxidation state changes of the photosystem II manganese complex in four transitions of the water oxidation cycle (S0 --> S1, S1 --> S2, S2 --> S3, and S3,4 --> S0) characterized by X-ray absorption spectroscopy at 20 K and room temperature, Biochemistry 44, 1894-1908. 44. Liang, W. C., Roelofs, T. A., Cinco, R. M., Rompel, A., Latimer, M. J., Yu, W. O., Sauer, K., Klein, M. P., and Yachandra, V. K. (2000) Structural change of the Mn cluster during the S2 -> S3 state transition of the oxygen- evolving complex of photosystem II. Does it reflect the onset of water/substrate oxidation? Determination by Mn X-ray absorption spectroscopy, J. Am. Chem. Soc. 122, 3399-3412. 45. Messinger, J., Wacker, U., and Renger, G. (1991) Unusual low reactivity of the water oxidase in redox state S3 toward exogenous reductants - analysis of the NH2OH-Induced and NH2NH2-induced modifications of flash- induced oxygen evolution in isolated spinach thylakoids, Biochemistry 30, 7852-7862. 46. Dau, H., Liebisch, P., and Haumann, M. (2005) The manganese complex of oxygenic photosynthesis: Conversion of five-coordinated Mn(III) to six- coordinated Mn(IV) in the S2-S3 transition is implied by XANES simulations, Physica Scripta T115, 844-846. 47. Iuzzolino, L., Dittmer, J., Dorner, W., Meyer-Klaucke, W., and Dau, H. (1998) X-ray absorption spectroscopy on layered photosystem II membrane particles suggests manganese-centered oxidation of the oxygen-evolving complex for the S0-S1, S1-S2, and S2-S3 transitions of the water oxidation cycle, Biochemistry 37, 17112-17119. 48. Messinger, J., Robblee, J. H., Bergmann, U., Fernandez, C., Glatzel, P., Visser, H., Cinco, R. M., McFarlane, K. L., Bellacchio, E., Pizarro, S. A., Cramer, S. P., Sauer, K., Klein, M. P., and Yachandra, V. K. (2001) Ab- sence of Mn-centered oxidation in the S2 -> S3 Transition: Implications for the mechanism of photosynthetic water oxidation, J. Am. Chem. Soc. 123, 7804-7820.

61 49. Dau, H., Liebisch, P., and Haumann, M. (2003) X-ray absorption spectroscopy to analyze nuclear geometry and electronic structure of biological metal centers - potential and questions examined with special focus on the tetra-nuclear manganese complex of oxygenic photosynthesis, Anal. Bio- anal. Chem. 376, 562-583. 50. Siegbahn, E. M. (2008) A structure-consistent mechanism for dioxygen formation in photosystem II, Chemistry-a European Journal 14, 8290- 8302. 51. Haumann, M., Liebisch, P., Muller, C., Barra, M., Grabolle, M., and Dau, H. (2005) Photosynthetic O2 formation tracked by time-resolved X-ray experiments, Science 310, 1019-1021. 52. Hwang, H. J., Dilbeck, P., Debus, R. J., and Burnap, R. L. (2007) Mutation of arginine 357 of the CP43 protein of photosystem II severely impairs the catalytic S-state cycle of the H2O oxidation complex, Biochemistry 46, 11987-11997. 53. Sproviero, E. M., Gascon, J. A., McEvoy, J. P., Brudvig, G. W., and Ba- tista, V. S. (2007) Quantum mechanics/molecular mechanics structural models of the oxygen-evolving complex of photosystem II, Current Opin- ion in Structural Biology 17, 173-180. 54. McEvoy, J. P., and Brudvig, G. W. (2006) Water-splitting chemistry of photosystem II, Chemical Reviews 106, 4455-4483. 55. Dau, H., and Haumann, M. (2007) Eight steps preceding O-O bond formation in oxygenic photosynthesis -a basic reaction cycle of the photosystem II manganese complex, Biochim. Biophys. Acta 1767, 472-483. 56. Debus, R. J., Strickler, M. A., Walker, L. M., and Hillier, W. (2005) No evidence from FTIR difference spectroscopy that aspartate-170 of the D1 polypeptide ligates a manganese ion that undergoes oxidation during the S0 to S1, S1 to S2, or S2 to S3 transitions in photosystem II, Biochemistry 44, 1367-1374. 57. Strickler, M. A., Walker, L. M., Hillier, W., and Debus, R. J. (2005) Evi- dence from biosynthetically incorporated strontium and FTIR difference spectroscopy that the C-terminus of the D1 polypeptide of photosystem II does not ligate calcium, Biochemistry 44, 8571-8577. 58. Strickler, M. A., Hillier, W., and Debus, R. J. (2006) No evidence from FTIR difference spectroscopy that glutamate-189 of the D1 polypeptide ligates a Mn ion that undergoes oxidation during the S0 to S1, S1 to S2, or S2 to S3 transitions in photosystem II, Biochemistry 45, 8801-8811. 59. Strickler, M. A., Walker, L. M., Hillier, W., Britt, R. D., and Debus, R. J. (2007) No evidence from FTIR difference spectroscopy that aspartate-342 of the D1 polypeptide ligates a Mn ion that undergoes oxidation during the S0 to S1, S1 to S2, or S2 to S3 transitions in photosystem II, Biochemistry 46, 3151-3160. 60. Strickler, M. A., Hwang, H. J., Burnap, R. L., Yano, J., Walker, L. M., Service, R. J., Britt, R. D., Hillier, W., and Debus, R. J. (2008) Glutamate- 354 of the CP43 polypeptide interacts with the oxygen-evolving Mn4Ca cluster of photosystem II: a preliminary characterization of the Glu354Gln mutant, Philos. Trans. R. Soc. B-Biol. Sci. 363, 1179-1187.

62 61. Allahverdiyeva, Y., Deak, Z., Szilard, A., Diner, B. A., Nixon, P. J., and Vass, I. (2004) The function of D1-H332 in Photosystem II electron transport studied by thermoluminescence and chlorophyll fluorescence in site- directed mutants of Synechocystis 6803, Euro. J. Biochem. 271, 3523- 3532. 62. Chu, H. A., Nguyen, A. P., and 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, 5839-5858. 63. McEvoy, J. P., and Brudvig, G. W. (2004) Structure-based mechanism of photosynthetic water oxidation, Physical Chemistry Chemical Physics 6, 4754-4763. 64. Ho, F. M., and Styring, S. (2008) Access channels and methanol binding site to the CaMn4 cluster in Photosystem II based on solvent accessibility simulations, with implications for substrate water access, Biochim. Biophys. Acta 1777, 140-153. 65. Murray, J. W., and Barber, J. (2007) Structural characteristics of channels and pathways in photosystem II including the identification of an oxygen channel, Journal of Structural Biology 159, 228-237. 66. Wydrzynski, T., Hillier, W., and Messinger, J. (1996) On the functional significance of substrate accessibility in the photosynthetic water oxidation mechanism, Physiol. Plant. 96, 342-350. 67. Ho, F. M. (2008) Uncovering channels in photosystem II by computer modelling: current progress, future prospects, and lessons from analogous systems, Photosynth. Res. 98, 503-522. 68. Messinger, J., Badger, M., and Wydrzynski, T. (1995) Detection of one slowly exchanging substrate water molecule in the S3 State of photosystem II, Proc. Natl. Acad. Sci. U.S.A. 92, 3209-3213. 69. Hillier, W., and Wydrzynski, T. (2000) The affinities for the two substrate water binding sites in the O-2 evolving complex of photosystem II vary in- dependently during S-state turnover, Biochemistry 39, 4399-4405. 70. Hillier, W., and Wydrzynski, T. (2008) 18O-water exchange in photosystem II: Substrate binding and intermediates of the water splitting cycle, Coordi- nation Chemistry Reviews 252, 306-317. 71. Hillier, W., and Wydrzynski, T. (2004) Substrate water interactions within the Photosystem II oxygen evolving complex, Physical Chemistry Chemi- cal Physics 6, 4882-4889. 72. Noguchi, T. (2008) FTIR detection of water reactions in the oxygenevolv- ing centre of photosystem II, Philos. Trans. R. Soc. Lond. B. Biol. Sci. 363, 1189-1195. 73. Suzuki, H., Sugiura, M., and Noguchi, T. (2008) Monitoring water reactions during the S-state cycle of the photosynthetic water-oxidizing center: Detection of the DOD bending vibrations by means of Fourier transform infrared spectroscopy, Biochemistry 47, 11024-11030. 74. Su, J. H., Lubitz, W., and Messinger, J. (2008) Probing mode and site of substrate water binding to the oxygen-evolving complex in the S2 state of

63 photosystem II by 17O-HYSCORE spectroscopy, J. Am. Chem. Soc. 130, 786-787. 75. Commoner, B., Heise, J. J., and Townsend, J. (1956) Light-induced para- magnetism in chloroplasts, Proc. Natl. Acad. Sci. U.S.A. 42, 710-718. 76. Barry, B. A., and Babcock, G. T. (1987) Tyrosine radicals are involved in the photosynthetic oxygen-evolving system, Proc. Natl. Acad. Sci. U.S.A. 84, 7099-7103. 77. Un, S., Atta, M., Fontecave, M., and Rutherford, A. W. (1995) g-values as a probe of the local protein environment - high-field EPR of tyrosyl radicals in ribonucleotide reductase and photosystem II, J. Am. Chem. Soc. 117, 10713-10719. 78. Un, S., Tang, X. S., and Diner, B. A. (1996) 245 GHz high-field EPR study of tyrosine-Do and tyrosine-Zo in mutants of photosystem II, Biochemistry 35, 679-684. 79. Hoganson, C. W., and Babcock, G. T. (1992) Protein tyrosyl radical interactions in photosystem-II studied by electron-spin-resonance and electron nuclear double-resonance spectroscopy - comparison with ribonucleotide reductase and in vitro tyrosine, Biochemistry 31, 11874-11880. 80. Rigby, S. E. J., Nugent, J. H. A., and Omalley, P. J. (1994) The dark stable tyrosine radical of photosystem-2 studied in 3 species using ENDOR and EPR spectroscopies, Biochemistry 33, 1734-1742. 81. Tommos, C., Tang, X. S., Warncke, K., Hoganson, C. W., Styring, S., McCracken, J., Diner, B. A., and Babcock, G. T. (1995) Spin-density distribution, conformation, and hydrogen-bonding of the redox-active tyrosine YZ in Photosystem II from multiple electron magnetic-resonance spectroscopies: Implications for photosynthetic oxygen evolution, J. Am. Chem. Soc. 117, 10325-10335. 82. Warncke, K., Babcock, G. T., and Mccracken, J. (1994) Structure of the YD tyrosine radical in photosystem-II as revealed by 2H electron-spin echo envelope modulation (ESEEM) spectroscopic analysis of hydrogen hyperfine interactions, J. Am. Chem. Soc. 116, 7332-7340. 83. Noguchi, T., Inoue, Y., and Tang, X. S. (1997) Structural coupling between the oxygen-evolving Mn cluster and a tyrosine residue in photosystem II as revealed by Fourier transform infrared spectroscopy, Biochemistry 36, 14705-14711. 84. Berthomieu, C., Hienerwadel, R., Boussac, A., Breton, J., and Diner, B. A. (1998) Hydrogen bonding of redox-active tyrosine Z of photosystem II probed by FTIR difference spectroscopy, Biochemistry 37, 10547-10554. 85. Hienerwadel, R., Boussac, A., Breton, J., Diner, B. A., and Berthomieu, C. (1997) Fourier transform infrared difference spectroscopy of photosystem II tyrosine D using site-directed mutagenesis and specific isotope labeling, Biochemistry 36, 14712-14723. 86. Hienerwadel, R., Boussac, A., Breton, J., and Berthomieu, C. (1996) Fou- rier transform infrared difference study of tyrosineD oxidation and plastoquinone QA reduction in photosystem II, Biochemistry 35, 15447-15460.

64 87. Diner, B. A., and Britt, R. D. (2005) The redox-active tyrosines YZ and YD, in Photosystem II: The light-driven water.plastoquinone oxidoreductase (Wydrzynski, T., and Satoh, K., Eds.) pp 207-233, Springer, Dordrecht. 88. Diner, B. A. (2001) Amino acid residues involved in the coordination 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. 89. Force, D. A., Randall, D. W., Britt, R. D., Tang, X. S., and Diner, B. A. (1995) 2H ESE-ENDOR study of hydrogen bonding to the tyrosine radicals YD' and Yz' of Photosystem II, J. Am. Chem. Soc. 117, 12643-12644. 90. Tang, X. S., Chisholm, D. A., Dismukes, G. C., Brudvig, G. W., and Diner, B. A. (1993) Spectroscopic evidence from site-directed mutants of Synechocystis PCC6803 in favor of a close interaction between histidine 189 and redox-active tyrosine 160, both of polypeptide D2 of the photosystem II reaction center, Biochemistry 32, 13742-13748. 91. Mamedov, F., Sayre, R. T., and Styring, S. (1998) Involvement of histidine 190 on the D1 protein in electron/proton transfer reactions on the donor side of photosystem II, Biochemistry 37, 14245-14256. 92. Hays, A. M. A., Vassiliev, I. R., Golbeck, J. H., and Debus, R. J. (1998) Role of D1-His190 in proton-coupled electron transfer reactions in photosystem II: A chemical complementation study, Biochemistry 37, 11352- 11365. 93. Eckert, H. J., and Renger, G. (1988) 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, 425-431. 94. Christen, G., Reifarth, F., and Renger, G. (1998) On the origin of the '35-μs kinetics' of P680+. reduction in photosystem II with an intact water oxidis- ing complex, FEBS Lett. 429, 49-52. 95. Karge, M., Irrgang, K. D., Sellin, S., Feinaugle, R., Liu, B., Eckert, H. J., Eichler, H. J., and Renger, G. (1996) Effects of hydrogen deuterium exchange on photosynthetic water cleavage in PS II core complexes from spinach, FEBS Lett. 378, 140-144. 96. Haumann, M., Bogershausen, O., Cherepanov, D., Ahlbrink, R., and Junge, W. (1997) Photosynthetic oxygen evolution: H/D isotope effects and the coupling between electron and proton transfer during the redox reactions at the oxidizing side of Photosystem II, Photosynth. Res. 51, 193-208. 97. Christen, G., and Renger, G. (1999) The role of hydrogen bonds for the +dot multiphasic P680 reduction by YZ in photosystem II with intact oxygen evolution capacity. Analysis of kinetic H/D isotope exchange effects, Bio- chemistry 38, 2068-2077. 98. Christen, G., Seeliger, A., and Renger, G. (1999) P680+dot reduction kinetics and redox transition probability of the water oxidizing complex as a function of pH and H/D isotope exchange in spinach thylakoids, Biochem- istry 38, 6082-6092. 99. Razeghifard, M. R., and Pace, R. J. (1997) Electron paramagnetic resonance kinetic studies of the S states in spinach PSII membranes, Biochim. Biophys. Acta 1322, 141-150.

65 100. Babcock, G. T., and Sauer, K. (1975) The rapid component of electron paramagnetic resonance signal II: a candidate for the physiological donor to photosystem II in spinach chloroplasts, Biochim. Biophys. Acta 376, 329- 44. 101. Hoganson, C. W., and Babcock, G. T. (1988) Electron-transfer events near the reaction center in O2-evolving photosystem-II preparations, Biochemis- try 27, 5848-5855. 102. Ruzicka, F. J., Beinert, H., Schepler, K. L., Dunham, W. R., and Sands, R. H. (1975) Interaction of ubisemiquinone with a paramagnetic component in heart tissue, Proc. Natl. Acad. Sci. U.S.A. 72, 2886-90. 103. Boussac, A., Zimmermann, J. L., and Rutherford, A. W. (1989) EPR signals from modified charge accumulation states of the oxygen evolving enzyme in Ca2+-deficient photosystem II, Biochemistry 28, 8984-8989. 104. Force, D. A., Randall, D. W., and Britt, R. D. (1997) Proximity of acetate, . manganese, and exchangeable deuterons to tyrosine YZ in acetate-inhibited . photosystem II membranes: implications for the direct involvement of YZ in water-splitting, Biochemistry 36, 12062-70. 105. Baumgarten, M., Philo, J. S., and Dismukes, G. C. (1990) Mechanism of photoinhibition of photosynthetic water oxidation by Cl- depletion and F- Substitution - oxidation of a protein residue, Biochemistry 29, 10814- 10822. 106. Sivaraja, M., Tso, J., and Dismukes, G. C. (1989) A calcium-specific site influences the structure and activity of the manganese cluster responsible for photosynthetic water oxidation, Biochemistry 28, 9459-9464. 107. Boussac, A., Zimmermann, J. L., Rutherford, A. W., and Lavergne, J. (1990) Histidine oxidation in the oxygen-evolving photosystem-II enzyme, Nature 347, 303-306. 108. Andreasson, L. E., and Lindberg, K. (1992) The inhibition of photosynthetic oxygen evolution by ammonia probed by EPR, Photosynth. Res. 34, 144-144. 109. Hallahan, B. J., Nugent, J. H. A., Warden, J. T., and Evans, M. C. W. (1992) Investigation of the origin of the S3 EPR signal from the oxygen- evolving complex of photosystem-2 - the role of tyrosineZ, Biochemistry 31, 4562-4573. 110. Maclachlan, D. J., and Nugent, J. H. A. (1993) Investigation of the S3 electron-paramagnetic-resonance signal from the oxygen-evolving complex of photosystem-2 - effect of inhibition of oxygen evolution by acetate, Bio- chemistry 32, 9772-9780. 111. Vanvliet, P., Boussac, A., and Rutherford, A. W. (1994) Chloride-depletion effects in the calcium-deficient oxygen-evolving complex of photosystem- II, Biochemistry 33, 12998-13004. 112. vanVliet, P., and Rutherford, A. W. (1996) Properties of the chloride- depleted oxygen-evolving complex of photosystem II studied by electron paramagnetic resonance, Biochemistry 35, 1829-1839. 113. Szalai, V. A., and Brudvig, G. W. (1996) Formation and decay of the S3 EPR signal species in acetate-inhibited photosystem II, Biochemistry 35, 1946-1953.

66 114. Szalai, V. A., Kuhne, H., Lakshmi, K. V., and Brudvig, G. W. (1998) Char- acterization of the interaction between manganese and tyrosine Z in acetate-inhibited photosystem II, Biochemistry 37, 13594-13603. 115. Dorlet, P., Di Valentin, M., Babcock, G. T., and McCracken, J. L. (1998) dot Interaction of YZ with its environment in acetate-treated photosystem II membranes and reaction center cores, Journal of Physical Chemistry B 102, 8239-8247. 116. Lydakis-Simantiris, N., Dorlet, P., Ghanotakis, D. F., and Babcock, G. T. dot 2+ (1998) Kinetic and spectroscopic properties of the YZ radical in Ca - and Cl--depleted photosystem II preparations, Biochemistry 37, 6427-6435. 117. Dorlet, P., Boussac, A., Rutherford, A. W., and Un, S. (1999) Multifre- quency high-field EPR study of the interaction between the tyrosyl Z radical and the manganese cluster in plant photosystem II, Journal of Physical Chemistry B 103, 10945-10954. 118. Lakshmi, K. V., Eaton, S. S., Eaton, G. R., Frank, H. A., and Brudvig, G. W. (1998) Analysis of dipolar and exchange interactions between manga- dot nese and tyrosine Z in the S2YZ state of acetate-inhibited photosystem II via EPR spectral simulations at X- and Q-Bands, Journal of Physical Chemistry B 102, 8327-8335. 119. Berthomieu, C., and Boussac, A. (1995) Histidine oxidation in the S2 to S3 transition probed by FTIR difference spectroscopy in the Ca2+-Depleted Photosystem-II - comparison with histidine radicals generated by UV irradiation, Biochemistry 34, 1541-1548. 120. Andreasson, L. E., Vass, I., and Styring, S. (1995) Ca2+ depletion modifies the electron-transfer on both donor and acceptor sides in photosystem-II from spinach, Biochim. Biophys. Acta 1230, 155-164. 121. Gilchrist, M. L., Ball, J. A., Randall, D. W., and Britt, R. D. (1995) Prox- imity of the manganese cluster of photosystem-II to the redox-active Tyro- sine YZ, Proc. Natl. Acad. Sci. U.S.A. 92, 9545-9549. 122. Tang, X. S., Randall, D. W., Force, D. A., Diner, B. A., and Britt, R. D. (1996) Manganese-tyrosine interaction in the photosystem II oxygen- evolving complex, J. Am. Chem. Soc. 118, 7638-7639. 123. Astashkin, A. V., Mino, H., Kawamori, A., and Ono, T. A. (1997) Pulsed EPR study of the S'3 signal in the Ca2+-depleted photosystem II, Chemical Physics Letters 272, 506-516. 124. Mino, H., Kawamori, A., Matsukawa, T., and Ono, T. (1998) Light-induced high-spin signals from the oxygen evolving center in Ca2+-depleted photosystem II studied by dual mode electron paramagnetic resonance spectroscopy, Biochemistry 37, 2794-2799. 125. Mino, H., Kawamori, A., and Ono, T. (2000) Pulsed EPR studies of doublet signal and singlet-like signal in oriented Ca2+-depleted PSII membranes: Location of the doublet signal center in PSII, Biochemistry 39, 11034- 11040. 126. Mino, H., and Itoh, S. (2005) EPR properties of a g=2 broad signal trapped 2+ in an S1 state in Ca -depleted photosystem II, Biochim. Biophys. Acta 1708, 42-49.

67 127. Ioannidis, N., Zahariou, G., and Petrouleas, V. (2006) Trapping of the S2 to S3 state intermediate of the oxygen-evolving complex of photosystem II, Biochemistry 45, 6252-6259. 128. Nugent, J. H., Muhiuddin, I. P., and Evans, M. C. (2002) Electron transfer from the water oxidizing complex at cryogenic temperatures: the S1 to S2 step, Biochemistry 41, 4117-4126. 129. Nugent, J. H., Turconi, S., and Evans, M. C. (1997) EPR investigation of water oxidizing photosystem II: detection of new EPR signals at cryogenic temperatures, Biochemistry 36, 7086-7096. 130. Zhang, C., and Styring, S. (2003) Formation of split electron paramagnetic resonance signals in photosystem II suggests that tyrosineZ can be photooxidized at 5 K in the S0 and S1 states of the oxygen-evolving complex, Biochemistry 42, 8066-8076. 131. Ioannidis, N., and Petrouleas, V. (2000) Electron paramagnetic resonance signals from the S3 state of the oxygen-evolving complex. A broadened radical signal induced by low-temperature near-infrared light illumination, Biochemistry 39, 5246-5254. 132. Zahariou, G., Ioannidis, N., Sioros, G., and Petrouleas, V. (2007) The collapse of the tyrosine Z*-Mn spin-spin interaction above approximately 100 K reveals the spectrum of tyrosine Z*. An application of rapid-scan EPR to the study of intermediates of the water splitting mechanism of photosystem II, Biochemistry 46, 14335-14341. 133. Dismukes, G. C., and Siderer, Y. (1981) Intermediates of a Polynuclear Manganese Center Involved in Photosynthetic Oxidation of Water, Proc. Natl. Acad. Sci. U.S.A. 78, 274-278. 134. Åhrling, K. A., Peterson, S., and Styring, S. (1998) The S0 state EPR signal from the Mn cluster in photosystem II arises from an isolated S = 1/2 ground state, Biochemistry 37, 8115-20. 135. Yamauchi, T., Mino, H., Matsukawa, T., Kawamori, A., and Ono, T. (1997) Parallel polarization electron paramagnetic resonance studies of the S1-state manganese cluster in the photosynthetic oxygen-evolving system, Biochemistry 36, 7520-7526. 136. Matsukawa, T., Mino, H., Yoneda, D., and Kawamori, A. (1999) Dual- mode EPR study of new signals from the S3-state of oxygen-evolving complex in photosystem II, Biochemistry 38, 4072-4077. 137. Sioros, G., Koulougliotis, D., Karapanagos, G., and Petrouleas, V. (2007) The S1YZ* metalloradical EPR signal of photosystem II contains two distinct components that advance respectively to the multiline and g = 4.1 con- formations of S2, Biochemistry 46, 210-217. 138. Ioannidis, N., Zahariou, G., and Petrouleas, V. (2008) The EPR spectrum of tyrosine Z* and its decay kinetics in O2-evolving photosystem II preparations, Biochemistry 47, 6292-6300. 139. Mamedov, F., Havelius, K. G. V., and Styring, S. (2008) in Photosysnthe- sis. Energy from the Sun: 14th international congress on photosysnthesis (Allen, J. F., Gantt, E., Golbeck, J. H., and Osmond, B., Eds.) pp 445-449, Springer, Glasgow.

68 140. Evelo, R. G., Hoff, A. J., Dikanov, S. A., and Tyryshkin, A. M. (1989) An ESEEM study of the oxidized electron-donor of plant photosystem-II - evidence that D. is a neutral tyrosine radical, Chemical Physics Letters 161, 479-484. 141. Zimmermann, J. L., Boussac, A., and Rutherford, A. W. (1993) The manganese center of oxygen-evolving and Ca2+-depleted photosystem .2. A pulsed EPR spectroscopy study, Biochemistry 32, 4831-4841. 142. Campbell, K. A., Peloquin, J. M., Diner, B. A., Tang, X. S., Chisholm, D. A., and Britt, R. D. (1997) The τ-nitrogen of D2 histidine 189 is the hydro- dot gen bond donor to the tyrosine radical YD of photosystem II, J. Am. Chem. Soc. 119, 4787-4788. 143. Koulougliotis, D., Teutloff, C., Sanakis, Y., Lubitz, W., and Petrouleas, V. (2004) The S1YZ metalloradical intermediate in photosystem II: an X- and W-band EPR study, Phys. Chem. Chem. Phys. 6, 4859-4863. 144. Davydov, R., Chemerisov, S., Werst, D. E., Rajh, T., Matsui, T., Ikeda- Saito, M., and Hoffman, B. M. (2004) Proton transfer at helium temperatures during dioxygen activation by heme monooxygenases, J. Am. Chem. Soc. 126, 15960-15961. 145. Hay, S., and Scrutton, N. S. (2008) H-transfers in Photosystem II: what can we learn from recent lessons in the enzyme community?, Photosynth. Res. 98, 169-77. 146. Faller, P., Goussias, C., Rutherford, A. W., and Un, S. (2003) Resolving intermediates in biological proton-coupled electron transfer: a tyrosyl radical prior to proton movement, Proc. Natl. Acad. Sci. U.S.A. 100, 8732- 8735. 147. Zhang, C., Boussac, A., and Rutherford, A. W. (2004) Low-temperature electron transfer in photosystem II: a tyrosyl radical and semiquinone charge pair, Biochemistry 43, 13787-13795. 148. Boussac, A., Kuhl, H., Ghibaudi, E., Rogner, M., and Rutherford, A. W. (1999) Detection of an electron paramagnetic resonance signal in the S0 state of the manganese complex of photosystem II from Synechococcus elongatus, Biochemistry 38, 11942-11948. 149. Styring, S. A., and Rutherford, A. W. (1988) The Microwave-Power Satu- ration of Signal IIslow Varies with the Redox State of the Oxygen-Evolving Complex in Photosystem-II, Biochemistry 27, 4915-4923. 150. Dexheimer, S. L., and Klein, M. P. (1992) Detection of a Paramagnetic Intermediate in the S1-State of the Photosynthetic Oxygen-Evolving Com- plex, J. Am. Chem. Soc. 114, 2821-2826. 151. Boussac, A., Girerd, J. J., and Rutherford, A. W. (1996) Conversion of the spin state of the manganese complex in photosystem II induced by near- infrared light, Biochemistry 35, 6984-6989. 152. Boussac, A., Un, S., Horner, O., and Rutherford, A. W. (1998) High-spin states (S >/= 5/2) of the photosystem II manganese complex, Biochemistry 37, 4001-4007. 153. Koulougliotis, D., Shen, J. R., Ioannidis, N., and Petrouleas, V. (2003) Near-IR irradiation of the S2 state of the water oxidizing complex of photo-

69 system II at liquid helium temperatures produces the metalloradical intermediate attributed to S1YZ*, Biochemistry 42, 3045-3053. 154. Ioannidis, N., Nugent, J. H., and Petrouleas, V. (2002) Intermediates of the S3 state of the oxygen-evolving complex of photosystem II, Biochemistry 41, 9589-9600. 155. Ioannidis, N., and Petrouleas, V. (2002) Decay products of the S3 state of the oxygen-evolving complex of photosystem II at cryogenic temperatures. Pathways to the formation of the S = 7/2 S2 state configuration, Biochemis- try 41, 9580-9588. 156. Sanakis, Y., Ioannidis, N., Sioros, G., and Petrouleas, V. (2001) A novel S = 7/2 configuration of the Mn cluster of photosystem II, J. Am. Chem. Soc. 123, 10766-10767. 157. Charlot, M. F., Boussac, A., and Blondin, G. (2005) Towards a spin coupling model for the Mn4 cluster in Photosystem II, Biochim. Biophys. Acta 1708, 120-132. 158. Horner, O., Riviere, E., Blondin, G., Un, S., Rutherford, A. W., Girerd, J. J., and Boussac, A. (1998) SQUID magnetization study of the infrared- induced spin transition in the S2 state of photosystem II: Spin value associated with the g = 4.1 EPR signal, J. Am. Chem. Soc. 120, 7924-7928. 159. Baxter, R., Krausz, E., Wydrzynski, T., and Pace, R. J. (1999) Identifica- tion of the near-infrared absorption band from the Mn cluster of photosystem II, J. Am. Chem. Soc. 121, 9451-9452. 160. Cooper, S. R., and Calvin, M. (1977) Mixed-valence interactions in di-μ- oxo bridged manganese complexes, J. Am. Chem. Soc. 99, 6623-6630. 161. Gamelin, D. R., Kirk, M. L., Stemmler, T. L., Pal, S., Armstrong, W. H., Pennerhahn, J. E., and Solomon, E. I. (1994) Electronic-structure and spectroscopy of manganese catalase and di-μ-oxo [Mn(III)Mn(IV)] model complexes, J. Am. Chem. Soc. 116, 2392-2399. 162. Boussac, A., Sugiura, M., Kirilovsky, D., and Rutherford, A. W. (2005) Near-infrared-induced transitions in the manganese cluster of photosystem II: action spectra for the S2 and S3 redox states, Plant Cell Physiol 46, 837- 842. 163. Deak, Z., Peterson, S., Geijer, P., Ahrling, K. A., and Styring, S. (1999) Methanol modification of the electron paramagnetic resonance signals from the S0 and S2 states of the water-oxidizing complex of photosystem II, Bio- chim. Biophys. Acta 1412, 240-9. 164. Pace, R. J., Smith, P., Bramley, R., and Stehlik, D. (1991) EPR saturation and temperature-dependence studies on signals from the oxygen-evolving center of photosystem II, Biochim. Biophys. Acta 1058, 161-170. 165. Zimmermann, J. L., and Rutherford, A. W. (1986) Electron paramagnetic resonance properties of the S2 state of the oxygen-evolving complex of photosystem-II, Biochemistry 25, 4609-4615. 166. Messinger, J., Nugent, J. H., and Evans, M. C. (1997) Detection of an EPR multiline signal for the S0* state in photosystem II, Biochemistry 36, 11055-60.

70 167. Åhrling, K. A., Peterson, S., and Styring, S. (1997) An oscillating manganese electron paramagnetic resonance signal from the S0 state of the oxygen evolving complex in photosystem II, Biochemistry 36, 13148-52. 168. Messinger, J., Robblee, J. H., Yu, W. O., Sauer, K., Yachandra, V. K., and Klein, M. P. (1997) The S0 state of the oxygen-evolving complex in photosystem II is paramagnetic: Detection of EPR multiline signal, J. Am. Chem. Soc. 119, 11349-11350. 169. Nöring, B., Shevela, D., Renger, G., and Messinger, J. (2008) Effects of methanol on the Si-state transitions in photosynthetic water-splitting, Pho- tosynth. Res. 98, 251-260. 170. Force, D. A., Randall, D. W., Lorigan, G. A., Clemens, K. L., and Britt, R. D. (1998) ESEEM studies of alcohol binding to the manganese cluster of the oxygen evolving complex of photosystem II, J. Am. Chem. Soc. 120, 13321-13323. 171. Åhrling, K. A., Evans, M. C. W., Nugent, J. H. A., Ball, R. J., and Pace, R. J. (2006) ESEEM studies of substrate water and small alcohol binding to the oxygen-evolving complex of photosystem II during functional turnover, Biochemistry 45, 7069-7082. 172. Schlodder, E., and Meyer, B. (1987) pH-dependence of oxygen evolution and reduction kinetics of photooxidized chlorophyll-aII (P-680) in photosystem II particles from Synechococcus sp., Biochim. Biophys. Acta 890, 23-31. 173. Meyer, B., Schlodder, E., Dekker, J. P., and Witt, H. T. (1989) O2 Evolu- + + tion and Chl-aII (P-680 ) nanosecond reduction kinetics in single flashes as a function of pH, Biochim. Biophys. Acta 974, 36-43. 174. Ahlbrink, R., Haumann, M., Cherepanov, D., Bogershausen, O., Mulkid- janian, A., and Junge, W. (1998) Function of tyrosine Z in water oxidation by photosystem II: electrostatical promotor instead of hydrogen abstractor, Biochemistry 37, 1131-1142. 175. Kühn, P., Pieper, J., Kaminskaya, O., Eckert, H. J., Lechner, R. E., Shu- valov, V., and Renger, G. (2005) Reaction pattern of Photosystem II: oxi- dative water cleavage and protein flexibility, Photosynth. Res. 84, 317-323. 176. Mulkidjanian, A. Y. (1999) Photosystem II of green plants: on the possible role of retarded protonic relaxation in water oxidation, Biochim. Biophys. Acta 1410, 1-6. 177. Schilstra, M. J., Rappaport, F., Nugent, J. H. A., Barnett, C. J., and Klug, D. R. (1998) Proton/hydrogen transfer affects the S state-dependent microsecond phases of P680+ reduction during water splitting, Biochemistry 37, 3974-3981. 178. Bernat, G., Morvaridi, F., Feyziyev, Y., and Styring, S. (2002) pH dependence of the four individual transitions in the catalytic S-cycle during photosynthetic oxygen evolution, Biochemistry 41, 5830-5843. 179. Suzuki, H., Sugiura, M., and Noguchi, T. (2005) pH dependence of the flash-induced S-state transitions in the oxygen-evolving center of photosystem II from Thermosynechoccocus elongatus as revealed by Fourier transform infrared spectroscopy, Biochemistry 44, 1708-1718.

71 180. Ono, T., and 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. 181. Faller, P., Debus, R. J., Brettel, K., Sugiura, M., Rutherford, A. W., and Boussac, A. (2001) Rapid formation of the stable tyrosyl radical in photosystem II, Proc. Natl. Acad. Sci. U.S.A. 98, 14368-14373. 182. Faller, P., Rutherford, A. W., and Debus, R. J. (2002) Tyrosine D oxidation at cryogenic temperature in photosystem II, Biochemistry 41, 12914-12920. 183. Hays, A. M. A., Vassiliev, I. R., Golbeck, J. H., and 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. 184. Ahlbrink, R., Semin, B. K., Mulkidjanian, A. Y., and Junge, W. (2001) 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, 117-126. 185. Geijer, P., Morvaridi, F., and Styring, S. (2001) The S3 state of the oxygen- evolving complex in photosystem II is converted to the S2YZ* state at alkaline pH, Biochemistry 40, 10881-10891. 186. Geijer, P., Deak, Z., and Styring, S. (2000) Proton equilibria in the manganese cluster of photosystem II control the intensities of the S0 and S2 state g ~ 2 electron paramagnetic resonance signals, Biochemistry 39, 6763-6772. 187. Tang, X. S., Zheng, M., Chisholm, D. A., Dismukes, G. C., and Diner, B. A. (1996) Investigation of the differences in the local protein environments dot dot surrounding tyrosine radicals YZ and YD in photosystem II using wild- type and the D2-Tyr160Phe mutant of Synechocystis 6803, Biochemistry 35, 1475-1484.

. / 0 1 2 / /

/ 2 / / 3/ . 3 // 4 5 / 6 7 5 / 4 35 / / 4 // 1

/ . / 1 2 / / 8 3993 5 4/

/: . /1 2 / / ; <

1 4 0 4/ 0400 0 .!==9=