Analysis of the role of detergent mixtures on the crystallization of the reaction center of Photosystem II

We have recently reported the crystallization of the reaction center of Photosystem II in the presence of detergent mixtures [Adir N (1999) Acta Crystallogr D Biol Crystallogr D55: 891–894]. We have used high performance liquid chromatography, dynamic light scattering, native gel electrophoresis and thermoluminescence measurements to characterize the interaction between these detergent mixtures and RC II, to try and understand their role in the crystallization process. Size exclusion HPLC and dynamic light scattering confirmed that the isolated RC II used for crystallization was exclusively monomeric. Dynamic light scattering measurements show that the detergent mixtures formed single micelles within a limited range of hydrodynamic radii. Both size exclusion HPLC and dynamic light scattering were used to follow the interaction between the detergent mixtures and monomeric RC II. These techniques revealed a decrease in the detergent mixture treated RC II particle size (with respect with the untreated RC II), and that RC II from solubilized crystals contained particles of the same size. Native gel electrophoresis showed that this change in apparent size is not due to the disintegration of the internal structure of the RC II complex. Thermoluminescence measurements of solubilized RC II crystals showed charge recombination from the S_2,3Q_A ⁻ state, indicating that RC II remains functionally viable following detergent mixture treatment and crystallization. The role of the detergent mixtures in the crystallization of RC II is discussed. This revised version was published online in June 2006 with corrections to the Cover Date.     

Molecular topology of the Photosystem II chlorophyll a binding protein,CP 43: Topology of a thylakoid membrane protein

We have used antibodies generated against synthetic peptides to determine the topology of the 43 kD chlorophyll a binding protein (CP 43) of Photosystem II. Based on the pattern of proteolytic fragments detected (on western blots) by peptide specific antibodies, a six transmembrane span topological model, with the amino and carboxyl termini located on the stromal membrane surface, is predicted. This structure is similar to that predicted for CP 47, a PS II chlorophyll a binding protein (Bricker T (1990) Photosynth Res 24: 1–13). The model is discussed in reference to the possible location of chlorophyll binding sites.This work was supported by National Institutes of Health Research Grant, GM40703 and U.S. Department of Energy Grant, DE-FG01-92ER20076 (to R.T.S.).     

Sequence similarity between Photosystems I and II. Identification of a Photosystem I reaction center transmembrane helix that is similar to transmembrane helix IV of the D2 subunit of Photosystem II and the M subunit of the non-sulfur purple and flexible green bacteria

There are basic structural similarities between plant PS II and bacterial RCs of the Chloroflexaceae and Rhodospirillaceae. These RCs are referred to as PS II-type RCs. A similar relationship of PS I RC to PS II-type RCs has not been established. Although plant PS I and PS II RCs show structural and functional differences, they also share similarities. Therefore, the A and B polypeptides of PS I were searched for PS II D1 and D2 polypeptide-like sequences. An alignment without gaps was found between PS II-type D2/M helix IV and PS I B helix X, as well as a weaker alignment of PS II-type D1/L with PS I B helix X. No comparable alignment with PS I A was found. In the M/D2 alignment there were eight identities and some conservative substitutions in twenty nine residues. PS I B helix X appeared to contain a modified chlorophyll dimer and monomer binding site and a modified non-heme iron-quinone binding site. The conserved residue sequence was found only in RC polypeptides. The proposed chlorophyll dimer-monomer binding site was located transmembrane from the iron-sulfur cluster X binding site. The conserved residues generally are those that interact with prosthetic groups. Half of the conserved residues are located on the same side of the helix. Thus, although there are impediments to concluding firmly that PS I B helix X has a functional and evolutionary relatedness to the D2 PS II and bacterial M RC polypeptides, our analysis gives reasonable support to the idea.Abbreviation RC reaction center     

Cloning, nucleotide sequence and mutational analysis of the gene encoding the Photosystem II manganese-stabilizing polypeptide of Synechocystis 6803

Summary Affinity purified, polyclonal antibodies raised against the Photosystem II 33 kDa manganese-stabilizing polypeptide of the spinach oxygen-evolving complex were used to isolate the gene encoding the homologous protein from Synechocystis 6803. Comparison of the amino acid sequence deduced from the Synechocystis psb1 nucleotide sequence with recently published sequences of spinach and pea confirms the homology indicated by antigenic crossreactivity and shows that the cyanobacterial and higher plant sequences are 43% identical and 63% conserved. Regions of identity, varying in length from 1 to 10 consecutive residues, are distributed throughout the protein. The 28 residues at the amino terminus of the psb1 gene product, characteristic of prokaryotic signal peptides, show homology with the carboxyl-terminal third of the transit sequences of pea and spinach and are most likely needed for the transport of the manganese-stabilizing protein across the thylakoid membrane to its destination of the lumen. Synechocystis mutants which contain a kanamycin resistance gene cassette inserted into the coding region for the 32 kDa polypeptide were constructed. These mutants contain no detectable 32 kDa polypeptide, do not evolve oxygen, and are incapable of photoautotrophic growth.     

On the role of the N-terminus of the extrinsic 33 kDa protein of Photosystem II

The role of the N-terminus of the extrinsic 33 kDa protein of Photosystem II has been investigated by means of site-directed mutagenesis and cross-linking. Replacement of Asp-9 resulted in a dramatic increase in proteolytic sensitivity leading to the degradation of the protein forming a 31 kDa fragment with an undefined N-terminus. This fragment was unable to restore oxygen evolution. However, the variants of the 33 kDa protein which remained intact could reconstitute oxygen evolution as effectively as the wild-type protein. Cross-linking experiments with a water-soluble carbodiimide revealed that mutagenesis of residue D9 led to the disruption of an intramolecular salt bridge. Therefore we suggest that the N-terminus of the 33 kDa protein is necessary for maintaining the binding ability of the protein to Photosystem II but might not be involved in binding itself.     

Functional Implications of Photosystem II Crystal Formation in Photosynthetic Membranes

The structural organization of proteins in biological membranes can affect their function. Photosynthetic thylakoid membranes in chloroplasts have the remarkable ability to change their supramolecular organization between disordered and semicrystalline states. Although the change to the semicrystalline state is known to be triggered by abiotic factors, the functional significance of this protein organization has not yet been understood. Taking advantage of an Arabidopsis thaliana fatty acid desaturase mutant (fad5) that constitutively forms semicrystalline arrays, we systematically test the functional implications of protein crystals in photosynthetic membranes. Here, we show that the change into an ordered state facilitates molecular diffusion of photosynthetic components in crowded thylakoid membranes. The increased mobility of small lipophilic molecules like plastoquinone and xanthophylls has implications for diffusion-dependent electron transport and photoprotective energy-dependent quenching. The mobility of the large photosystem II supercomplexes, however, is impaired, leading to retarded repair of damaged proteins. Our results demonstrate that supramolecular changes into more ordered states have differing impacts on photosynthesis that favor either diffusion-dependent electron transport and photoprotection or protein repair processes, thus fine-tuning the photosynthetic energy conversion.     

Control of Photosystem II in spinach leaves by continuous light and by light pulses given in the dark

The light-induced induction of components of non-photochemical quenching of chlorophyll fluorescence which are distinguished by different rates of dark relaxation (qNf, rapidly relaxing and qNs, slowly relaxing or not relaxing at all in the presence brief saturating light pulses which interrupt darkness at low frequencies) was studied in leaves of spinach.After dark adaptation of the leaves, a fast relaxing component developed in low light only after a lag phase. Quenching increased towards a maximum with increasing photon flux density. This fast component of quenching was identified as energy-dependent quenching qE. It required formation of an appreciable transthylakoid pH and was insignificant when darkened spinach leaves received 1 s pulses of light every 30 s even though zeaxanthin was formed from violaxanthin under these conditions.Another quenching component termed qNs developed in low light without a lag phase. It was not dependent on a transthylakoid pH gradient, decayed exponentially with a long half time of relaxation and was about 20% of total quenching irrespective of light intensity. When darkened leaves were flashed at frequencies higher than 0.004 Hz with 1 s light pulses, this quenching also appeared. Its extent was very considerable, and it did not require formation of zeaxanthin. Relaxation was accelerated by far-red light, and this acceleration was abolished by NaF.We suggest that qNs is the result of a so-called state transition, in which LHC II moves after its phosphorylation from fluorescent PS II to nonfluorescent PS I. This state transition was capable of decreasing in darkened leaves the potential maximum quantum efficiency of electron flow through Photosystem II by about 20%.Abbreviations PFD photon flux density - PS photosystem     

Structure and Activity of the Photosystem II Manganese-Stabilizing Protein: Role of the Conserved Disulfide Bond

The 33-kDa manganese-stabilizing protein (MSP) of Photosystem II (PS II) maintains the functional stability of the Mn cluster in the enzyme’s active site. This protein has been shown to possess characteristics similar to those of the intrinsically disordered, or natively unfolded proteins [Lydakis-Simantiris et al. (1999b) Biochemistry 38: 404–414]. Alternately it was proposed that MSP should be classified as a molten globule, based in part on the hypothesis that its lone disulfide bridge is necessary for structural stability and function in solution [Shutova et al. (2000) FEBS Lett. 467: 137–140]. A site-directed mutant MSP (C28A,C51A) that eliminates the disulfide bond reconstitutes O₂ evolution activity and binds to MSP-free PS II preparations at wild-type levels [Betts et al. (1996) Biochim. Biophys. Acta 1274: 135–142]. This mutant was further characterized by incubation at 90 °C to determine the effect of loss of the disulfide bridge on MSP thermostability and solution structure. After heating at 90 °C for 20 min, C28A,C51A MSP was still able to bind to PS II preparations at molar stoichiometries similar to those of WT MSP and reconstitute O₂ evolution activity. A fraction of the protein aggregates upon heating, but after resolubilization, it regains the ability to bind to PS II and reconstitute O₂ evolution activity. Characterization of the solution structure of C28A,C51A MSP, using CD spectroscopy, UV absorption spectroscopy, and gel filtration chromatography, revealed that the mutant has a more disordered solution structure than WT MSP. The disulfide bond is therefore unnecessary for MSP function and the intrinsically disordered characteristics of MSP are not dependent on its presence. However, the disulfide bond does play a role in the solution structure of MSP in vivo, as evidenced by the lability of a C20S MSP mutation in Synechocystis 6803 [Burnap et al. (1994) Biochemistry 33: 13712–13718].     

Current perceptions of Photosystem II

In the last few years our knowledge of the structure and function of Photosystem II in oxygen-evolving organisms has increased significantly. The biochemical isolation and characterization of essential protein components and the comparative analysis from purple photosynthetic bacteria (Deisenhofer, Epp, Miki, Huber and Michel (1984) J Mol Biol 180: 385–398) have led to a more concise picture of Photosystem II organization. Thus, it is now generally accepted that the so-called D1 and D2 intrinsic proteins bind the primary reactants and the reducing-side components. Simultaneously, the nature and reaction kinetics of the major electron transfer components have been further clarified. For example, the radicals giving rise to the different forms of EPR Signal II have recently been assigned to oxidized tyrosine residues on the D1 and D2 proteins, while the so-called Q₄₀₀ component has been assigned to the ferric form of the acceptor-side iron. The primary charge-separation has been meaured to take place in about 3 ps. However, despite all recent major efforts, the location of the manganese ions and the water-oxidation mechanism still remain largely unknown. Other topics which lately have received much attention include the organization of Photosystem II in the thylakoid membrane and the role of lipids and ionic cofactors like bicarbonate, calcium and chloride. This article attempts to give an overall update in this rapidly expanding field.     

Reconstitution of the spinach oxygen-evolving complex with recombinant Arabidopsis manganese-stabilizing protein

The psbO gene of cyanobacteria, green algae and higher plants encodes the precursor of the 33 kDa manganese-stabilizing protein (MSP), a water-soluble subunit of photosystem II (PSII). Using a pET-T7 cloning/expression system, we have expressed in Escherichia coli a full-length cDNA clone of psbO from Arabidopsis thaliana. Upon induction, high levels of the precursor protein accumulated in cells grown with vigorous aeration. In cells grown under weak aeration, the mature protein accumulated upon induction. In cells grown with moderate aeration, the ratio of precursor to mature MSP decreased as the optical density at induction increased. Both forms of the protein accumulated as inclusion bodies from which the mature protein could be released under mildly denaturing conditions that did not release the precursor. Renatured Arabidopsis MSP was 87% as effective as isolated spinach MSP in restoring O₂ evolution activity to MSP-depleted PSII membranes from spinach; however, the heterologous protein binds to spinach PSIIs with about half the affinity of the native protein. We also report a correction to the previously published DNA sequence of Arabidopsis psbO (Ko et al., Plant Mol Biol 14 (1990) 217–227).     

Light-Harvesting Features Revealed by the Structure of Plant Photosystem I

Oxygenic photosynthesis is driven by two multi-subunit membrane protein complexes, Photosystem I and Photosystem II. In plants and green algae, both complexes are composed of two moieties: a reaction center (RC), where light-induced charge translocation occurs, and a peripheral antenna that absorbs light and funnels its energy to the reaction center. The peripheral antenna of PS I (LHC I) is composed of four gene products (Lhca 1-4) that are unique among the chlorophyll a/b binding proteins in their pronounced long-wavelength absorbance and in their assembly into dimers. The recently determined structure of plant Photosystem I provides the first relatively high-resolution structural model of a super-complex containing a reaction center and its peripheral antenna. We describe some of the structural features responsible for the unique properties of LHC I and discuss the advantages of the particular LHC I dimerization mode over monomeric or trimeric forms. In addition, we delineate some of the interactions between the peripheral antenna and the reaction center and discuss how they serve the purpose of dynamically altering the composition of LHC I in response to environmental pressure. Combining structural insight with spectroscopic data, we propose how altering LHC I composition may protect PS I from excessive light.     

The extrinsic 33 kDa polypeptide of the oxygen-evolving complex of photosystem II is a putative calcium-binding protein and is encoded by a multi-gene family in pea

The extrinsic 33 kDa polypeptide of the water-oxidizing complex has been extracted from pea photosystem II particles by washing with alkaline-Tris and purified by ion-exchange chromatography. The N-terminal amino acid sequence has been determined, and specific antisera have been raised in rabbits and used to screen a pea leaf cDNA library in gt11. Determination of the nucleotide sequence of positive clones revealed an essentially full-length cDNA for the 33 kDa polypeptide, the deduced amino acid sequence showing it to code for a mature protein of 248 amino acids with an N-terminal transit peptide of 81 amino acids. The protein showed a high degree of conservation with previously reported sequences for the 33 kDa protein from other species and the sequence contained a putative Ca²⁺-binding site with homology to mammalian intestinal calcium-binding proteins. Northern analysis of total pea RNA indicated a message of approximately 1.4 kb, in good agreement with the size of the cDNA obtained at 1.3 kbp. Southern blots of genomic DNA probed with the labelled cDNA give rise to several bands suggesting that the 33 kDa polypeptide is coded by a multi-gene family.Abbreviations ATZ - anilinothiazolinone - DITC - p-phenylenediisothiocyanate - PTH - phenylthiohydantoin - TFA - trifluoroacetic acid - Tris - tris (hydroxymethyl) aminomethane - bis-Tris - bis (2-hydroxyethyl) imino-tris (hydroxymethyl)-methane - p.f.u. - plaque-forming units     

Crystallization of the chaperone protein SecB.

The secretory protein SecB found in Escherichia coli is a molecular chaperone that binds to precursor forms of a number of proteins targeted for export to the periplasmic space. SecB maintains these proteins in a translocation-competent conformation facilitating the translocation process. The material has been cloned and expressed in E. coli. Crystals have been grown from polyethylene glycol 8000 by vapor diffusion using the hanging drop technique. These crystals are monoclinic, belonging to space group C2 with unit cell dimensions a = 56.0 A, b = 111.1 A, c = 134.7 A, and beta = 104 degrees. The crystals diffract to 8 A resolution on a Rigaku imaging plate detector. Dynamic light scattering experiments suggest that SecB exhibits aggregation behavior with a number of different precipitating agents. These results may explain resistance of SecB to forming ordered crystals.     

Turnover of the D1 protein and of Photosystem II in a Synechocystis 6803 mutant lacking Tyrz

The Photosystem II reaction center is rapidly inactivated by light, particularly at higher light intensity. One of the possible factors causing this phenomenon is the oxidized primary donor, P680⁺, which may be harmful to Photosystem II because of its highly oxidizing nature. However, no direct evidence specificially implicating P680⁺ in photoinhibition has been obtained yet. To investigate whether P680⁺ is harmful to Photosystem II, turnover of the D1 protein and of the Photosystem II reaction center complex were measured in vivo in a mutant of the cyanobacterium Synechocystis sp. PCC 6803, in which the physiological donor to P680⁺, Tyr_z, was genetically deleted. In this mutant, D1 degradation in the light is an order of magnitude faster than in wild type. The most straightforward explanation of this phenomenon is that accumulation of P680⁺ leads to an increased rate of turnover of the Photosystem II reaction center complex, which is compatible with the hypothesis of destructive oxidation by P680⁺ that is damaging to the Photosystem II complex.     

The Arabidopsis PsbO2 protein regulates dephosphorylation and turnover of the photosystem II reaction centre D1 protein

The extrinsic photosystem II (PSII) protein of 33 kDa (PsbO), which stabilizes the water-oxidizing complex, is represented in Arabidopsis thaliana (Arabidopsis) by two isoforms. Two T-DNA insertion mutant lines deficient in either the PsbO1 or the PsbO2 protein were retarded in growth in comparison with the wild type, while differing from each other phenotypically. Both PsbO proteins were able to support the oxygen evolution activity of PSII, although PsbO2 was less efficient than PsbO1 under photoinhibitory conditions. Prolonged high light stress led to reduced growth and fitness of the mutant lacking PsbO2 as compared with the wild type and the mutant lacking PsbO1. During a short period of treatment of detached leaves or isolated thylakoids at high light levels, inactivation of PSII electron transport in the PsbO2-deficient mutant was slowed down, and the subsequent degradation of the D1 protein was totally inhibited. The steady-state levels of in vivo phosphorylation of the PSII reaction centre proteins D1 and D2 were specifically reduced in the mutant containing only PsbO2, in comparison with the mutant containing only PsbO1 or with wild-type plants. Phosphorylation of PSII proteins in vitro proceeded similarly in thylakoid membranes from both mutants and wild-type plants. However, dephosphorylation of the D1 protein occurred much faster in the thylakoids containing only PsbO2. We conclude that the function of PsbO1 in Arabidopsis is mostly in support of PSII activity, whereas the interaction of PsbO2 with PSII regulates the turnover of the D1 protein, increasing its accessibility to the phosphatases and proteases involved in its degradation.     

The mechanism of the degradation of psaB gene product,one of the photosynthetic reaction center subunits of Photosystem I,upon photoinhibition

The psaB gene product (PsaB protein), one of the reaction center subunits of Photosystem I (PS I), was specifically degraded by light illumination of spinach thylakoid membranes. The degradation of the protein yielded N-terminal fragments of molecular mass 51 kDa and 45 kDa. The formation of the 51 kDa fragment was i) partially suppressed by the addition of phenylmethylsulfonyl fluoride or 3,4-dichloroisocoumarin, which are inhibitors of serine proteases, and ii) enhanced in the presence of hydrogen peroxide during photoinhibitory treatment, but iii) not detected following hydrogen peroxide treatment in the dark. These results suggest that the hydroxyl radical produced at the reduced iron-sulfur centers in PS I triggers the conformational change of the PS I complex, which allows access of a serine-type protease to PsaB. This results in the formation of the 51 kDa N-terminal fragment, presumably by cleavage on the loop exposed to the stromal side, between putative helices 8 and 9. On the other hand, the formation of the 45 kDa fragment, which was enhanced in the presence of methyl viologen but did not accompany the photoinhibition of PS I, was not affected by the addition of hydrogen peroxide or protease inhibitors. Another fragment of 18 kDa was identified as a C-terminal counterpart of the 45 kDa fragment. N-terminal sequence analysis of the 18 kDa fragment revealed that the cleavage occurred between Ala⁵⁰⁰ and Val⁵⁰¹ on the loop exposed to the lumenal side, between putative helices 7 and 8 of the PsaB protein.     

Nitrogen ligation to the manganese cluster of Photosystem II in the absence of the extrinsic proteins and as a function of pH

Three extrinsic proteins (PsbO, PsbP and PsbQ), with apparent molecular weights of 33, 23 and 17 kDa, bind to the lumenal side of Photosystem II (PS II) and stabilize the manganese, calcium and chloride cofactors of the oxygen evolving complex (OEC). The effect of these proteins on the structure of the tetramanganese cluster, especially their possible involvement in manganese ligation, is investigated in this study by measuring the reported histidine-manganese coupling [Tang et al. (1994) Proc Natl Acad Sci USA 91: 704–708] of PS II membranes depleted of none, two or three of these proteins using ESEEM (electron spin echo envelope modulation) spectroscopy. The results show that neither of the three proteins influence the histidine ligation of manganese. From this, the conserved histidine of the 23 kDa protein can be ruled out as a manganese ligand. Whereas the 33 and 17 kDa proteins lack conserved histidines, the existence of a 33 kDa protein-derived carboxylate ligand has been posited; our results show no evidence for a change of the manganese co-ordination upon removal of this protein. Studies of the pH-dependence of the histidine–manganese coupling show that the histidine ligation is present in PS II centers showing the S₂ multiline EPR signal in the pH-range 4.2–9.5. This revised version was published online in June 2006 with corrections to the Cover Date.     

An FTIR study on the structure of the oxygen-evolving Mn-cluster of Photosystem II in different spin forms of the S2 state

The S₂ state of the oxygen-evolving Mn-cluster of Photosystem II (PS II) is known to have different forms that exhibit the g =2 multiline and g = 4.1 EPR signals. These two spin forms are interconvertible at > 200 K and the relative amplitudes of the two signals are dependent on the species of cryoprotectant and alcohol contained in the medium. Also, it was recently found that the mutiline form can be converted to the g = 4.1 form by absorption of near-infrared light by the Mn-cluster itself at around 150 K [Boussac et al. (1996) Biochemistry 35: 6984–6989]. We have used light-induced Fourier transform infrared (FTIR) difference spectroscopy to study the structural difference in these two S₂ forms. FTIR difference spectra for S₂/S₁ as well as for S₂Q_A ^-/S₁Q_A measured at cryogenic temperatures using PS II membranes in the presence of various cryoprotectants, and monohydric alcohols did not show any specific differences except for intensities of amide I bands, which were larger when ethylene glycol or glycerol was present in addition to sucrose. This result was interpreted due to more flexible movement of the protein backbones upon S₂ formation with a higher cryoprotectant content. Light-induced difference spectra measured at 150 K using either blue light without near-infrared light or red plus near-infrared light also did not show any detectable difference. In addition, a different spectrum upon near-infrared illumination at 150 K of the PS II sample in which the S₂ state had been photogenerated at 200 K exhibited no meaningful signals. These results indicate that the two S₂ forms that give rise to the multiline and g = 4.1 signals have only minor differences, if any, in the structures of amino-acid ligands and polypeptide backbones. This conclusion suggests that conversion between the two spin states is caused by a spin-state transition in the Mn(III) ion rather than valence swapping within the Mn-cluster that would considerably affect the vibrations of ligands.This revised version was published online in October 2005 with corrections to the Cover Date.     

Oxygen evolution by Photosystem II: The contribution of backward transitions to the anomalous behaviour of double-hits revealed by a new analysis method

Backward transitions in the analysis of oxygen production under flashing light were introduced by Packham et al., 1988, Photosynth. Res. 15: 221–232. In order to take backward transitions into account, a new method of analysis is presented: the eigenvalue method. This method is based on the recurrence relation of oxygen production with four coefficients (also known as the four sigma coefficients). It shows less susceptibility to round-off errors than other methods and permits the computation of double-hits directly from the coefficients, which was not possible before. With it we discovered that the inconsistent behaviour of double-hits observed previously under low flash intensities or low flash frequencies was mainly due to the inclusion of the backward transitions into the double-hit probability. In these conditions backward transitions seemed to be due either to the combination of an S-state deactivation and a miss, or to two S-state deactivations and a single-hit.In the presence of 3-(3, 4-Dichlorophenyl)-1, 1-dimethylurea (DCMU), the previous methods of sigma analysis failed. In contrast, the new method resolved all four S-state probabilities; thus it has the further advantage of being more robust (robustness being defined as the ability to yield a meaningful answer under difficult conditions).Abbreviation: DCMU 3-(3, 4-Dichlorophenyl)-1, 1-dimethylurea     

Reaction pattern of Photosystem II: oxidative water cleavage and protein flexibility

This short communication addresses three topics of photosynthetic water cleavage in Photosystem II (PS II): (a) effect of protonation in the acidic range on the extent of the ‘fast’ ns kinetics of P680^+· reduction by Y_Z, (b) mechanism of O–O bond formation and (c) role of protein flexibility in the functional integrity of PS II. Based on measurements of light-induced absorption changes and quasielastic neutron scattering in combination with mechanistic considerations, evidence is presented for the protein acting as a functionally active constituent of the water cleavage machinery, in particular, for directed local proton transfer. A specific flexibility emerging above a threshold of about 230 K is an indispensable prerequisite for oxygen evolution and plastoquinol formation.