Absence of the psbH gene product destabilizes photosystem II complex and bicarbonate binding on its acceptor side in Synechocystis PCC 6803 Josef Komenda 1,2 , Lenka Lupõ  nkova  1,2 and Jir Ï õ  Kopecky  1,2 1 Photosynthesis Research Centre, University of South Bohemia, C Ï eske  Bude Ï jovice, Czech Republic; 2 Laboratory of Photosynthesis, Institute of Microbiology, Academy of Sciences, Tr Ï ebon Ï , Czech Republic The PsbH protein, a small subunit of the photosystem II complex (PSII), was identi®ed as a 6-kDa protein band in the PSII core and subcore (CP47±D1±D2±c yt b-559) from the wild-type s train of the cyanobacterium Synechocystis PCC 6803. The protein was missing in the D 1±D2±cyto- chrome b-559 complex and also in all PSII complexes isolated from IC7, a mutant lacking the psbH gene. The following properties of PSII in the mutant contrasted with those in wild-type: (a) CP47 was released during nondena- turing electrophoresis of the PSII core isolated from IC7; (b) depletion of CO 2 resulted in a reversible decrease of the Q À A reoxidation rate in the IC7 cells; (c) light-induced decrease in PSII activity, measured as 2,5±dimethyl-benzo- quinone-supported Hill reaction, was strongly dependent on the H CO 3 ± concentration i n t he IC7 cells; and (d) illumina- tion of the IC7 cells lead to an extensive oxidation, frag- mentation and cross-linking of the D 1 p rotein. We did not ®nd any evidence for phosphorylation of the PsbH protein in thewild-typestrain. The results showed that in the PSII complex of Synechocysti s attachment of CP47 to the D1±D2 heterodimer appears weakened a nd binding of bicarbonate on the PSII acceptor side is destabilized in the absence of the PsbH protein. Keywords: cyanobacteria; D1 protein; photosystem II; psbH gene; Synechocystis PCC 6803. The core of photosystem II (PSII) complex of higher plants, algae and cyanobacteria consists of large central subunits D1, D2, CP47, CP43 a nd a number of low molecular m ass proteins. It i s b elieved t hat w ith a n e xception of cyto- chrome b-559, the small proteins do not part icipate directly in the transfer of electrons within PSII but they are important for the optimization of electron transfer processes and for the proper assembly of the complex (reviewed in [1]). Two different strategies are used to get information about the r ole of small PSII subunits. One approach is based on the functional co mparison of the intact and detergent treated PSII complex missing a speci®c subunit. Resump- tion of the particular function after reconstitution of the complex with this subunit is considered as evidence for its role. F or example, using this approach the role o f the PsbL subunit i n t he Q A binding has been proposed [2]. The second, more frequent approach is based on the deletion of the gene encoding a studied protein followed by a detailed characterization of PSII complex in the resulting mutant. This strategy has been very often successful in cyanobac- teria, namely in the strain Synechocystis PCC 6803, which can grow photoheterotrophically and is easily transform- able. In this way, mutants of Synechocystis with deleted psbK, psbH, psbI [3±5] and other genes were constructed. These mutants contained assembled PSII complexes and after their functional characterization possible functions were ascribed to these subunits. Interestingly, in algae and higher plants this approach is useful only rar ely as deletions of PSII subunits usually lead to disappearance of the whole PSII complex from thylakoids [6±8]. The PsbH protein, a product of the psbH gene, w as initially found as the 10-kDa phosphoprotein i n thyla- koids of higher plants by Bennett [9]. From that time its homologues have been found in more than 15 photo- synthetic o rganisms including cyanobacteria. T he ®rst partial sequence of the cyanobacterial PsbH protein was obtained in the thermophilic cyanobacterium Synecho- coccus vulcanus [10], but the complete gene was sequenced in the strain Synechocystis P CC 6803 by Abdel-Mawgood & Dilley [11] and Mayes & Barber [12]. Construction of the Synechocystis psbH-less mutant and its characterization in vivo provided the ®rst more solid basis for the elucidation of the role of the protein in PSII [4]. The mutant was more sensitive to photoinhibition in comparison with the wild-type [4,13] and this sensitivity has been mostly attributed to perturbations in the electron ¯ow between Q A and Q B on the acceptor side of PSII. In the present paper we have conducted a more detailed analysis of the effects of the PsbH absence on the structure and function of PSII both in vivo and in vitro. Our results indicated a stabilizing role of the protein for CP47 binding to the D1±D2 heterodimer and showed its importance for bicarbonate binding and preventing oxidative stress in PSII. Correspondence to J. Komenda, Institute of Microbiology, Opatovicky mly n, 379 81 Tr Ï ebon Ï , Czech Republic. Fax: + 420 333 721246, Tel.: + 420 333 721101, E-mail: komenda@alga.cz Abbreviations: cyt, cytochrome; DM, dodecylmaltoside; DCBQ, 2,5-dichloro-p-benzoquinone; DMBQ, 2,5-dimethyl-p-benzoquinone; DNP-, dinitrophenyl-; HRA, Hill reaction activity; PSI and PSII, photosystem I and photosystem II complexes; PSII RC, reaction centre complex of photosystem II; ROS, reactive oxygen species. (Received 1 5 August 2001 , revised 16 November 2001, accepted 20 November 2 001) Eur. J. Biochem. 269, 610±619 (2002) Ó FEBS 2002 MATERIALS AND METHODS Strains and growth of organisms The glucose tolerant strain Synechocystis PCC 6803 [14], referred to as w ild-type (WT), and i ts psbH deletion mutant IC7 [4] were grown in BG-11 medium with (photomixo- trophic growth) or without (photoautotrophic growth) glucose ( 10 m M ®nal c oncentration). The plate medium contained BG-11, 10 m M Tes/NaOH,pH8.2,1.5%agar and 0.3% sodium thiosulphate [15] and in the case of the IC7 mutant also kanamycin (25 lgámL )1 ) and atrazine (5.10 )6 M ) were added. Liquid cultures (100±200 mL) in conical ¯asks were aerated using an orbital shaker, irradi- ated with 50±70 lmol photonsám )2 ás )1 of white light at 29 °C and diluted every day to maintain the chlorophyll concentration at % 8 lgámL )1 .CulturesofChlorella soro- kiniana, Scenedesmus quadricauda and Chla mydomonas reinhardtii were grown under the same conditions and their density was maintained at D 753 %1. Photoinhibitory treatment of the Synechocystis cultures was performed at a chlorophyll concentration of 6 lgámL )1 in 18-mm thick plate-parallel cuvettes placed in a temper- ature controlled bath. Cultures were bubbled with air containing 2% CO 2 (CO 2 -enriched air), with air bubbled through 40% NaOH (CO 2 -depleted air) or with pure nitrogen. In some experiments, the cell suspension was supplemented with 10 m M NaHCO 3 . The light source was a 500-W tungsten ®lament bulb mounted in an aluminium re¯ector. I n t he experim ents w ith a protein-synthesis inhibitor lincomycin (Sigma, USA, 100 lgámL )1 ®nal concentration) the culture was incubated for 10 min in the dark before the start of light treatment. Phosphorylation of membrane proteins in algal and Synechocystis strains was induced in the cell suspensions diluted to D 753  0.2. They were exposed to 250 lmol pho- tonsám )2 ás )1 of white light for 30 min either in the absence or in the presence of 3 lCiámL )133 P-H 3 PO 4 . Thylakoids isolated from the cells were analysed by SDS/PAGE and Western blotting using rabbit polyclonal antiphosphothre- onine antibody (Zymed, USA) or by a utoradiography. Preparation of membranes and PSII complexes and their trypsinization Cyanobacterial membranes were prepared by breaking the cells with glass beads (150±200 lmindiameter)at4°C followed by differential centrifugation. For small scale preparation, the cells (approx. 150 lg of chlorophyll) were washed and resuspended in 150 lLof25m M Tris/HCl buffer, pH 7.5 containing 1 m M aminocaproic acid. The beads were added to the suspension and the mixture was vortexed twice for 1 min with 2 min interruption for cooling on ice. Beads were then washed four times with 200 lLof buffer. Aliquots were pooled and centrifuged at 3000 g for 1 min to remove unbroken cells. Membranes were collected from the supernatant at 20 000 g for 10 min The ®nal sediment was resuspended in 25 m M Tris/HCl buffer, pH 6.8 containing 1 M sucrose (®nal chlorophyll concen- tration 400±600 lgámL )1 ) and stored at )75 °C. Large scale preparation of m embranes for isolation of PSII was performed a ccording to T ang & Diner [16] u sing a beadbeater (Biospec Products, USA) f or breaking the cells. Isolation of PSII complexes f rom the wild-type and mutant thylakoids was c onducted a ccording to the modi®ed procedure of Ritter et al. [17]. Brie¯y, membranes were spun down, resuspended in 25 m M Mes/NaOH, pH 6.5 and solubilized with dodecylmaltoside (DM/chlorophyll  20, w/w) for 15 min. Unsolubilized material was removed by centrifugation (40 000 g, 15 min). The supernatant was applied on the column of chelating Sepharose (Amersham Pharmacia, Sweden) with bound Cu 2+ ions and imidazole equilibrated with t wo column volumes of 25 m M Mes/ NaOH, pH 6.5 containing 200 m M NaCl and 0.03% DM. PSII and carotenoid fraction did not bind to the column and went through directly into the second column of Q Sepha- rose (Amersham Pharmacia, Sweden). This was washed with several volumes of 25 m M Mes/NaOH, pH 6.5 con- taining 200 m M NaCl and 0.03% DM. During this step, carotenoids and remaining small amounts of phycocyano- biliproteins and PSI were removed from the column. Finally, t he PSII core complex was eluted from the column by 25 m M Mes/NaOH, pH 6.5 containing 250 m M NaCl and 0.03% DM. The preparation was concentrated in Centricon 30 spin columns (Millipore, USA). Trypsinization of membranes was performed at chloro- phyll concentration 2 00 lgámL )1 and trypsin co ncentration 50 lgámL )1 (Serva, Germany). After 5 , 15 and 30 min incubation at 25 °C, aliquots were withdrawn and proteo- lysis was stopped by transfer to ice and addition of 2 m M Pefabloc SC (Merck, Germany). Analysis of proteins Isolated PSII complexes or membranes solubilized with DM (DM/chlorophyll  20, w/w) were analysed by nondena- turing electrophoresis at 4 °C in 5±10% polyacrylamide gel according to Laemmli [18] except that the electrophoretic buffers contained 12.5 m M Tris, 98 m M glycine and 0.1% Deriphat 160, and the gel contained 0.1 M Tris/HCl, pH 8.8 without detergent. Protein composition of membranes and p igment protein complexes obtained by Deriphat electrophoresis was assessed by electrophoresis in a d enaturing 12±20% linear gradient polyacrylamide gel containing 7 M urea [18]. The membranes were solubilized in 25 m M Tris/HCl, pH 6.8, containing 2% SDS (w/v) and 2% dithiothreitol (w/v) at laboratory temperature for 60 min Samples were loaded with equal amount of chlorophyll as indicated in ®gure legends. Analysis o f pigment proteins was p erformed either by re-electrophoresis of individual pigment protein bands or the whole lane from the native gel was excised and placed on the top of the SDS gel (diagonal PAGE). The gels with p igment proteins were incubated for one hour in the same solubilization solution as thylakoids prior to SDS/PAGE. Proteins separated in the gel were either stained by C oomassie Blue or tran sferred onto nitrocellulose membrane (0.1 lm, Schleicher-Schuel, Ger- many) by semidry blotting. M embrane was incubated with speci®c antibodies and then with alkaline phosphatase conjugated secondary antibody (Sigma). Proteins were visualized by colorimetric reac tion using B CPIP-NBT system. Antibodies used in the study were raised against: (a) residues 2±17 of the Synechocystis PCC 6803 D1 protein (D1-Nt); ( b) residues 5 8±86 o f the s pinach D1 protein (D1-Mp); (c) the last 29 residues of the pea Ó FEBS 2002 Function of the PsbH protein in photosystem II (Eur. J. Biochem. 269) 611 D1 precursor (D1-Ct); (d) the last 14 residues of the Synechocystis D2 (D2-Ct) and (e) the isolated a subunit of the cytochrome b-559 from Synechocystis PCC 6803 (cyt b-559). For autoradiography, the membrane with labelled proteins was exposed to X-ray ®lm at laboratory temperature for 2 days. Oxidation of proteins Oxidation of the D1 protein was determined using t he detection kit Oxyblot (Intergen, USA). Solubilized thyla- koid m embrane proteins w ere derivatized using dini- trophenylhydrazine, which reacts with carbonyls present on oxidized proteins. After protein separation b y SDS/ PAGE and transfer onto the membrane, dinitrophenyl (DNP)-proteins were detected by Western blotting using anti-DNP Ig. The whole procedure was performed accord- ing to manufacturer's instructions. N-terminal protein sequencing N-terminal sequence of proteins was analysed performing eight cycles of a utomated Edman degradations using Protein sequencer LF3600D (Beckman, USA) and program 2±39 according to manufacturer's instructions. A mino-acid sequence was called from the comparisons of chromato- grams. Protein in the gel was blotted onto poly(vinylidene di¯uoride) membrane, prewetted with acetonitrile, and then deblocked by treatment with 0.6 M HCl for 20 h at 25 °C. HCl was then evaporated and the membrane was inserted into cartridge of the sequencer. Measurement of oxygen evolution Light-saturated steady-state rates of oxy gen evolution (Hill reaction activity, HRA) in cell suspensions were measured at 3 0 °C u sing a t emperature controlled chamber [19] equipped w ith a Clark-type electrode (YSI, USA). Arti®cial e lectron acceptors 2 ,5-dimethyl-p-benzoquinone (DMBQ) or 2,6-dichloro-p-benzoquinone (DCBQ) (0.5 m M ®nal concentration each) were added 1 min before the measuring illumination (3500 lmol photonsám )2 ás )1 ,30s) was switched on. Chlorophyll ¯uorescence measurement TherateofQ À A reoxidation was measured with the P.S.I. double-modulated ¯uorometer FL-100 (P.S.I., Czech republic). Short, nonactivating pulses of blue light were used as the m easuring light and F M re¯ecting fully reduced Q A was elicited by the strong saturating red ¯ash. Cells were incubated for 5 min in the dark before measurements. Pigment analyses For the routine measurements of chlorophyll concentration, the cells were collected by centrifugation and extracted w ith 100% methanol. The concentration of chlorophyll was calculated from the absorbance values of the extract at 666 and 720 nm according to Wellburn and Lichtenthaler [20]. Detailed analysis of pigments was performed by HPLC (Beckman, USA) using procedure of Gilmore and Yama- moto [2 1]. RESULTS Identi®cation of the PsbH protein in photosystem II complexes of Synechocystis PSII core complexes from wild-type and IC7 strains of Synechocyst is were isolated by a combination of metal af®nity and ionex chromatography. Absorption spectra of preparations from each strain exhibited similar absorption maxima at 673 nm, typical for P SII complex from this cyanobacterial species [16]. The preparations were then subjected to the nondenaturing electrophoresis in the presence of Deriphat 160 (Fig. 1). In the case of wild-type, we obtained two prominent green bands (Fig. 1A). The ®rst band was ascribed to the monomeric PSII core consisting of CP47, CP43, D2, D1, both cytochrome b-559 subunits, a 6-kDa protein and other smaller proteins (Fig. 1B, WT: A). The s econd band represented PSII core lacking CP43 (PSII subcore, Fig. 1B, WT: B). There were also two low molecular mass pigment-containing bands that were ascribed to free CP43 based on its protein composition (Fig. 1B, WT: D) and free carotenoids (Fig. 1A, FP) based on its absorption spectrum (data not shown). Similar electrophoretic pattern of the pigment proteins was obtained from the IC7 strain with the exception that: (a) the band of the PSII subcore was much weaker than in wild- type (b) there was an additional band identi®ed as the D1± D2±cyt b-559 complex (PSII RC) (Fig. 1B, IC7: C), and (c) the lower green band contained both CP47 and CP43 (Fig. 1B, IC7: D). The results suggest that during electro- phoresis the PSII subcore from IC7 became unstable and decomposed into CP47 and PSII RC c omplex. Comparison of the protein composition of the PSII cores and subcores (Fig. 1B) from both strains showed that there was a protein with M r of % 6 kDa in the complexes from wild-type that was absent in IC7. The band was subjected to the automated Edman degradation. The obtained sequence DILRPLNS corresponding to the internal sequence 8±15 of the PsbH protein from Synechocystis PCC 6803 (SWISS- PROT accession number P14835) con®rmed t he identity of the protein. Analysis of protein composition of PSII complexes from wild-type revealed that PsbH protein was present in the core as well as in the subcore complex lack ing CP43. Evaluation of its presence in PSIIRC was allowed by the treatment of the wild-type preparation with SDS in the ratio SDS/ chlorophyll  10. Deriphat PAGE of this preparation led to the generation of PSIIRC that was devoid of the PsbH protein (Fig. 2). It means that this subunit was released from PSII subcore together with CP47, again suggesting a close structural relationship between PsbH and C P47. Effect of the PsbH absence on the accessibility of the D1 protein to trypsin The e ffect of the PsbH absence on the structure of the PSII core complex was further probed by trypsinization of the D1 protein in isolated membranes of wild-type and IC7 (Fig. 3). The initial trypsin-induced cut of the D1 protein occurred at the N-terminus and was documented by a small increase of the electrophoretic mobility and by a loss of reactivity with the D1-Nt Ig (data not shown). As in Synechococc us PCC 7942, this cutting occurred concomi- 612 J. Komenda et al. (Eur. J. Biochem. 269) Ó FEBS 2002 tantly with breakdown of the D2 protein at residue R234 and there was also trypsin-induced formation of the 35-kDa adduct of t he D2 C-terminal fragment and D1 protein without N-terminus [22]. Interestingly, formation of this D1±D2 adduct was inhibited in the IC7 mutant. After these initial events the D1 protein was subsequently cut at residue K238 and later also at R257 generating the C -terminal 12 and 10-kDa Ct1 and Ct2 fragments. A 20-kDa Nt1 fragment reacting with D1-Mp and very weakly with D1-Nt represented the D1 subfragment between residues R8 and K238. However, in the IC7 mutant, a 16-kDa Nt2 fragment reacting with D1-Mp Ig a lso appeared. As j udged from the amino-acid sequence of the protein, the Nt2 fragment originated from the cut at R225. In summary, trypsinization of thylakoids showed that in the absence of the PsbH protein the accessibility of the D1 protein to trypsin was changed and also mutual position between D1 and D2 was modi®ed a s i ndicated b y i nhibition of the D1± D2 adduct formation in the I C7 strain. The PsbH protein affects the bicarbonate binding on the acceptor side of PSII A characteristic feature of PSII in the IC7 strain is a slow electron transfer between Q A and Q B [4]. This was demonstrated in the Fig. 4 (compare solid lines in the left and right panels). Additional retardation of the electron transfer could be induced by removal of CO 2 from the medium during dark incubation of the mutant cells. This retardation was fully reversed after subsequent addition of bicarbonate and /or bubbling with the CO 2 -enriched air (Fig. 4, right panel). In contrast, removal of CO 2 and its subsequent addition did not affect the Q À A reoxidation rate in the wild-type strain (Fig. 4, left panel). It indicated that the binding of bicarbonate to the PSII acceptor side was weakened in the IC7 mutant as a consequence of the missing PsbH protein. This conclusion was supported by the following experiment. We have shown previously that after exposure to high irradiance, the Hill reaction activity of the IC7 cells measured using 2,5-dimethyl-benzoquinone as an arti®cial electron acceptor (DMBQ-HRA) was very quickly inhibited. In contrast, the decline of activity measured using 2,6-dichloro-benzoquinone (DCBQ-HRA), was much slower [13]. We found that this difference was further enhanced when the illuminated IC7 cells were bubbled with the CO 2 -depleted air (Fig. 5, closed symbols). However, when the suspension was supplemented with 5 m M bicar- bonate and bubbled with CO 2 -enriched air, the decline of DMBQ- and DCBQ-HRA was parallel (Fig. 5, open symbols). The rate of DMBQ- and DCBQ-HRA decline in the wild-typ e cells was not dependent on the C O 2 and/or bicarbonate concentration (not shown, see [ 13]). The D1 protein is extensively photooxidized in the mutant It was shown previously that the turnover of the D1 protein in the illuminated IC7 cells is retarded and also the recovery from photoinhibition is slow as compared with the wild- type cells [13]. Possible explanation for this feature of the IC7 s train could be a n increased formation o f r eactive oxygen species (ROS) in PSII that m ay inhibit t he D1 replacement process [23]. The ®rst supporting evidence for this came from the analysis of pigment content in the autotrophically grown wild-type and IC7. We assumed that increased formation of ROS could lead to increase of cellular carotenoid content, as these pigments are able to eliminate to some extent the ROS effect. HPLC analysis revealed almost four times higher ratio of myxoxantho- phyll/chlorophyll in the IC7 cells as compared to the wild- type cells (Table 1). The increase in content of other carotenoids was not as signi®cant. Increased generation of Fig. 1. Identi®cation of the PsbH protein by analysis of PSII complexes isolated from wild-type and IC7 mutant. (A) Pigment protein pro®le of PSII complexes isolated from wild-type and IC7 after the native PAGE in the presence o f Deriphat 160, 8 lg o f chlorophyll loade d per l ane. (B) Protein composition of pigment proteins obtained by native Deri- phat/PAGE of PSII complexes isolated from wild-type and IC7: A, PSII core  PSII complex containing at least CP47, CP43, D1, D2 and cytochrome b-559; B, PSII subcore  PSII core comp lex lac king CP43; C, PSIIRC  D1±D2±cytochrome b-559 complex; D, CP47, C P43  free pigment p roteins CP47 and CP43; and F P  free pigments. Ó FEBS 2002 Function of the PsbH protein in photosystem II (Eur. J. Biochem. 269) 613 ROS in the PSII complex of IC7 was further supported b y the r esults of the D1 analysis in cells exposed to high irradiance. The Western blot showed, in addition to the typical 32-kDa D 1 b and, formation of a 40-kDa band that also reacted with the antibody raised against the a subunit of cytochrome b-559 (Fig. 6). Although this band was present even in control cells, high irradiance induced formation of an a dditional, slightly smaller D1±cyto- chrome b-559-reactive band. We propose that this band was identical to that found by Barbato et al. [24] in illuminated plant thylakoids which seems to be induced by the action of ROS [25]. Effect of high irradiance was further accompanied by decreased i ntensity of the original 32-k Da band and in the case of IC7 mobility of the remaining protein was decr eased in an oxygen-dependent manner. Such a shift often re¯ects protein oxidation [26] and this was con®rmed by O xyblot, a co mmercially available kit devel- oped to detect oxidized proteins. Indeed, after light treat- ment of IC7 cells the D1 protein with lower mobility exhibited signi®cant oxidation that was partially inhibited in the cells bubbled with nitrogen during illumination. In addition, a 23-kDa N-terminal D1 fragment was detected in the IC7 cells and its mobility was also shifted by high irradiance in the presence of oxygen. As showed by Miyao [25], also fragmentation of the D1 protein may be induced by ROS. Taken together, the above results provide strong experimental support for enhanced generation of ROS in the PSII complex lacking the PsbH protein. The PsbH of Synechocystis is not phosphorylated in vivo The PsbH protein has originally been identi®ed in higher plants due to its phosphorylation in light [9]. This phosphorylation also exists in green algae and two N-terminal threonine residues seem to be phosphorylated in these organisms [27,28]. However, th e question concern- ing phosphorylation of the cyanobacterial PsbH remains still open. There is a single report documenting in vitro phosphorylation of PsbH in Synechocystis by Race & Gounaris [29]. However, in this report identi®cation of the phosphorylated ban d as the PsbH protein was ambiguous as in thylakoids there is a dozen of polypeptides below Fig. 2. The PsbH protein is absent in the PSII RC complex o f wild-type. PSII complex of wild-type isolated by chromatography has been analysed after an addition of SDS in the ratio (w/w) SDS/chlorophyll  1or10bythe native PAGE in the presence of Deriphat 160 in the ®rst dimension (8 lg of chlorophyll loaded per lane) and SDS/PAGE in t he sec- ond dimension according t o Materials and methods (diagonal PAGE). Fig. 3. Time course of the D1 trypsinolysis in membranes isolated from wild-type and IC7 cells. Isolated membranes were incubated with trypsin and samples were taken at indicated time intervals for elec- trophoresis of proteins and immunoblotting as described in Materials and methods (5 lg of chlorophyll loaded per lane). Nitrocellulose membrane with separated thylakoid pr oteins was probed with a mix- ture of anti-(D1-Mp) Ig and anti-(D1-Ct) Ig according to Materials and methods. Nt1  fragment of D1 between R8 a nd K238; Nt2  fragment of D1 between R8 and R225; Ct1  fragmentofD1 between K238 and A344; Ct2  fragment of D1 between R257 and A344; a nd D1D2 adduct  adduct between D 1 (R8-A344) and frag- ment of D2 (R234-L352). 614 J. Komenda et al. (Eur. J. Biochem. 269) Ó FEBS 2002 10 kDa. In addition, the cyanobacterial PsbH lacks the N-terminal sequence with threonine 2 and 4 residues phosphorylated in plants and algae [30,31]. To clarify this point, we attempted to id entify phosphorylation of the PsbH in vivo using antiphosphothreonine Ig that proved to react well with PSII phosphoproteins in plant chloroplasts [32]. For comparison, we also analysed PsbH phosphory- lation in green algae Scenedesmus quadricauda, Chlorella sorokiniana and Chlamydomonas reinhardtii. A single 6.5- kDa band in Scenedesmus and two closely migrating 8 and 9-kDa bands in Chlorella and Chlamydomonas could be detected in the PSII cores (Fig. 7A). In contrast, absolutely no reaction of cyanobacterial proteins with the antibody in membranes and PSII core complexes of both wild-type and IC7 suggested that the threonine phosphorylation does not occur in thylakoids of Synechocystis. In order to extend this conclusion for phosphorylation of other residues, the cells of Synechocystis were labelled with 33 P-H 3 PO 4 . We found weak phosphorylation of two bands with M r values of 3 and 4 kDa clearly distinct from the PsbH protein (Fig. 7B). In addition, th ese two bands were p resent in both wild-type and the IC7 mutant. In summary, we did not obtain any experimental evidence for the phosphorylation of the PsbH protein in Synechocyst is cells. DISCUSSION Packham [33] proposed that the PsbH protein of photo- system II is a functional homologue of the H subunit from the reaction centre of photosynthetic bacteria. This is in line with the effect of the protein on the herbicide binding in PSII [34] and a lso in line with the data in this paper. The r ecent 3.8 A Ê model o f P SII [35] tentatively s ituated the membrane helix of PsbH on the side o f CP47 a nd D2 in the proximity of the Q A ±Q B region. This position provides a good justi®cation for the stabilizing effect of PsbH o n the binding of CP47 to the heterodimer D1±D2 as well as on the bicarbonate binding to the acceptor side of PSII. Never- theless, there is also recent report showing the PsbH protein in Chlamydomonas on the periphery of the PSII dimeric core [36]. However, detection of the protein was based on the attachment of gold particles to His-tagged N-terminus that can be positioned at the different region of the core than the membrane helix. In Synechocystis, the PsbH protein was found not only in the PSII core but also in the subcore (CP47±D1±D2±cyt) Fig. 4. Eect of CO 2 depletion on t he reoxi- dation of t he reduced PSII primary electron acceptor Q A in wild-type and IC7 cells. Cul- tures of wild-type (left p anel) and IC7 (right panel) gr own in the presence of glucose were bubbled in the d ark at 30 °Cwith CO 2 -enriched air for 3 0 min (solid lines), then with CO 2 -depleted air for another 30 min (dotted line) and ®nally again w ith CO 2 - enriched air f or 30 min (dashed line). The kinetics of the Q A reoxidation was me asured by P.S.I. ¯uorometer as described in Materials and methods. Fig. 5. Eect of CO 2 on the DMBQ-HRA and DCBQ-HRA during illumination of IC7 cells. Culturegrowninthepresenceofglucosewas bubbled with CO 2 -depleted (closed symbols) or CO 2 -enriched air (open symbols) a t 30 °C for 10 min in the dark and then during illu- mination with 1000 lmo l photonsám )2 ás )1 , aliquots of cells were taken at the times indicated for measurement of DMBQ-HRA (circles) and DCBQ-HRA (squares). Culture bubbled w ith C O 2 -enriched a ir co n- tained 10 m M bicarbonate in addition. Means of a t least three mea- surements are shown, s.e. did not exceed 8%. T he initial values of DMBQ-HRA and DCBQ-HRA were 180  30 and 270  40 lmol O 2 mg (chlorophyll) )1 h )1 , respectively. Table 1. Carotenoid composition in cells of wild-type and IC7 strains grown in the absence of glucose. Numbers represent the percentage of the total carotenoids, numbers in parenthesis represent the percentage of the particular carotenoids taking content per chlorophyll unit in wild-type cells as 100%. Myxoxanthophyll Zeaxanthin Echinenon b-caroten Wild-type cells 22 (100) 35 (100) 19 (100) 23 (100) IC7 cells 47 (347) 23 (107) 13.5 (115) 16 (115) Ó FEBS 2002 Function of the PsbH protein in photosystem II (Eur. J. Biochem. 269) 615 complex. In contrast, the PsbH protein was not detected in the subcore isolated from spinach [37]. The reason for this discrepancy is unclear, but it could be related to the difference either between the species or between the methods of the subcore preparation. The instability of t he IC7 s ubcore during electrophoresis of the isolated PSII core can be relevant to the situation in vivo during the PSII core assembly. Weak binding of CP47 to the D1±D2 heterodimer may destabilize these subunits to the extent that they are degraded before the whole c omplex can be assembled. This proteolysis seems to be less ef®cient in cyanobacteria than that in algae and therefore the assembly of PSII complexes occurs in the psbH-deletion mutant of Synechocystis, but not in the similar mutant of Chlamydomonas [6,7]. On the other hand, the PsbH protein may also represent an important factor regulating process of PSII repair. Its removal from PSII could result in a complete disassembly of PSII during the D1 replacement w hile in its p resence the D1 replacement could proceed in the subcore complex as suggested by Zhang et al. [38]. We have identi®ed formation of the D1±cyto- chrome b-559 adduct and the D1 fragments together with the apparent oxidation of the D1 protein in the cells of IC7. This shows that the impaired function of PSII in IC7 leads to increased probability of the formation of ROS. These species oxidize the D1 protein which can be subsequently cross-linked with the a subunit of cytochrome b-559, or even fragmented. However, ROS may also attack other PSII subunits as well as protein synthesis apparatus and then the r ecovery from photoinh ibition i s slow a s observed in IC7 [13]. Oxidative damage was also implicated in the slow restoration of PSII activity after photoinhibition of Synechocyst is [39] and Synechococcus elongatus cells [23]. We were not able to accelerate r ecovery from photoinhibi- tion in IC7 by bubbling the cell suspension with nitrogen during high irradiance treatment. Nevertheless, this does not negate our hypothesis as even under these conditions, oxidation of the D1 protein still occurred although to a lesser extent (Fig. 6). Importance o f t he PsbH protein for the proper f unction- ing of the PSII complex in h igher plants and algae is c losely related to phosphorylation of its threonine residues on the N-terminus. However, in Synechocystis we did not ®nd any evidence for the phosphorylation of this protein. Looking at the N-terminal s equences of PsbH in organisms containing phycobilisomes attached to the stromal side of the mem- brane (e.g. Synechocystis, Synechococcus, Porphyra and Cyanidium), it is apparent that they contain the PsbH protein with shorter N -terminal part w ithout the phosph o- rylable threonines. As the common feature of these organ- isms is the absence of grana, it is possible that the Fig. 6. Oxidation , cross-linking and fragmen- tation of the D1 protein during illumination of the IC7 cells. Cells of wild-type and IC7 grown in the presence of g lucose were illuminated with 1000 lmol photonsám )2 ás )1 for 90 min, and after breaking the cells m embrane proteins were analysed by SDS/PAGE and Western blotting. (A) Degradation and o x i- dation of D1: D1  content of the 32-kDa D1 band [anti-(D1-Mp) I g], 0.5 lg of c hloro- phyll loaded per lane; Oxy D1  oxidation of the 32-kDa D1 band ( an ti-DNP Ig), 5 lgof chlorophyll loaded per lane. (B) Fragmenta- tion and cross-linking of D1, 5 lg of c hloro- phyll loaded per lane; D1  32 kDaD1band; D1fr  N-terminal 23-kDa D1 f ragment;. D1ad, cyt ad  41 kDa D1±a cyto- chrome b-559 double band; cyt  a-subunit of cytochrome b-559. 616 J. Komenda et al. (Eur. J. Biochem. 269) Ó FEBS 2002 phosphorylation of PsbH is important for the function of PSII in the Se appressed regions of the membrane. In line with this, G iardi et al. [40] showed that af ter PsbH dephosphorylation by alkaline phosphatase an extremely fast inactivation of the PSII activity occurred in isolated spinach membranes while the phosphatase had n o effect on the activity of the cyanobacterial membranes. Xiong et al. [41] postulated a hypothesis suggesting that arginine residues o f the D1 protein (especially Arg257) stabilize b inding of bicarbonate on the PSII acceptor side. However, similar role could be ful®lled by arginines of the PsbH protein as suggested by Sundby et al. [42]. They found that phosphorylation of the PsbH protein is indirectly proportional to t he binding of bicarbonate on the acceptor side of PSII. Based on this correlation they proposed that the basic residues on the stromal side of the PsbH protein are i nvolved in the bicarbonate binding. From this point of view it is interesting that our results indicated destabilization of the bicarbonate binding in PSII as a consequence of the missing PsbH protein. However, it is not clear if the protein binds bicarbonate directly or whether it has long-distance effect on the conformation of D1 and/or D2 that is important for the binding of this anion. It is worth to note that fast light-induced inactivation of DMBQ-HRA, which most probably re¯ects release of bicarbonate, has been also found in the PEST-deletion mutant of Synechocystis by Nixon et al. [43]. It may i ndicate that the PEST sequence of the D1 protein is in close c ontact with PsbH and also contributes to the formation of the bicarbonate binding site. In line with this hypothesis our trypsinization experiment showed that in the absence of the PsbH protein the P EST region of the D1 p rotein was more exposed to stroma. The fact that the release of bicarbonate completely inhib ited the DMBQ-HRA but only slowed down the DCBQ-HRA suggests that DCBQ may accept electrons before the bicarbonate binding site. Therefore, it is tempting to speculate that the difference between the active (Q B -reducing) and inactive (Q B -nonreducing) PSII centres, having distinct af®nity to DMBQ and DCBQ [44], is given by the occupancy of the bicarbonate binding site and/or the state of the PsbH protein (e.g. phosphorylation) that affects the bicarbonate binding. ACKNOWLEDGEMENTS This work was supported by the projects no. LN00A141 of The Ministry of Education, You th and Sports of the C zech Republic, 203/00/1257 of the Grant Agency of the Czech Republic and also by Institutional Research Concept no . AV0Z5020903. W e thank Dr K. Bezous Ï ka for the sequencing of the Psb H protein, Ms. Jana Hofhanzlova  forableassistanceandProf.J.Ma  lek for critical reading of the manuscript. We are grateful to Prof. J. Barber for a kind gift of the IC7 mutant as well as Prof. E M. Aro, Dr Peter Nixon, Dr A. Mattoo and Dr R. B arbato who donated speci®c antisera. REFERENCES 1. Hankamer, B., Barber, J . & Boekema, E.J. (1997 ) Structure and membrane organization of p hotosystem-II in green plants. Annu. Rev. Plant Physiol. Plant Mol. Biol. 48, 641±671. 2. Kitamura, K., Ozawa, S., Shiina, T. & Toyoshima, Y. (1994) L protein, e ncoded by psbL, restores normal functioning of t he primary quino ne acce ptor, Q(A), in isolated D1/D2/CP47/Cytb- 559/I photosystem II reaction center c ore complex. FEBS Lett. 354, 113±116. Fig. 7. Analysis of phosphorylation of the PsbH protein by Western blot using antiphosphothreonine antibodies (left panel) and by radioactive labelling with 33 P-H 3 PO 4 (right panel). (A) Western blot. Membranes isolated from illuminated cells of green algae Chlorella (Chlo), Scenedesmus (Scen) and Chlamydomonas (Chla), and from Synechocystis strains w ild-type and IC7 were solubilized with DM and subjected to PAGE in the presence o f Deriphat 160 (8 lg of chlorophyll loaded per lane.). PSII core complexes were excised from the gel and further analysed by SDS/PAGE. Separated proteins were transferre d o nto n it rocellul ose mem brane w hich was subsequently used f or immu nodetec tion u sing an tiphospho threonine rabbit polyclonal antibodies (Zymed). (B) Autoradiogram. Membranes isolated from wild-type and IC7 cells were labelled with 33 P-H 3 PO 4 and analysed by SDS/PAGE (5 lg of chlorophyll loaded per lane). Separated proteins were transferred onto nitrocellulose membrane, which was subsequently used for autoradiography. Ó FEBS 2002 Function of the PsbH protein in photosystem II (Eur. J. Biochem. 269) 617 3. Ikeuchi, M., Eggers, B., Shen, G.Z., Webber, A., Yu J.J., Hirano, A., Inoue, Y. & Vermaas, W. (1991) Cloning of the psbK gene from Synechocystis sp. PCC-6803 and characterization of photosystem-II in mutants lacking PSII-K. J. Biol. Chem. 266, 11111±11115. 4. Mayes, S.R., Dubbs, J.M., Vass, I., Hideg, E., Nagy, L. & Barber, J. (1993) Further characterization of the psbH locus of Synechocystis sp. PCC-6803 ± inactivation o f psbH impairs Q(A) to Q(B) electron-transport in photosystem-2. Biochemistry 32, 1454±1465. 5. Ikeuchi, M., Shukla, V.K., Pakrasi, H.B. & Inoue, Y. (1995) Directed inactivation of the psbI gene do es not a ect photosystem - II in the cyanobacterium Synechocystis sp PCC 6803. Mol. General Genet. 249, 6 22±628. 6. Summer, E.J., Schmid, V.H.R., Bruns, B.U. & Schmidt, G.W. (1997) Requirement f or the H phosphoprotein in photosystem II of Chlamydomona s reinhardtii. Plant Ph ysiol. 113, 1 359±1368. 7. O'Connor, H.E., Rue, S.V., Cain, A.J., Deak, Z., Vass, I., Nugent, J.H.A. & Purton, S. (1998) The 9-kDa phosphoprotein of photosystem II. Generation and characterisation of Chlamydo- monas mutant lacking PSII-H and a site-directed mutant lacking the phosphorylation site. Biochim. Biophys. Acta 1364, 6 3±72. 8. Takahashi, Y., Matsumoto, H., Goldschmidt-Clermont, M. & Rochaix, J.D. (1994) Dire cted disruption of the Chlamydomonas chloroplast psbK gene destabilizes the photosystem II reaction- center co mplex. Plant. Mol. B iol. 24, 7 79±788. 9. Bennet, J. (1977) Phosphorylation of c hloroplast membrane polypeptides. Nature 269, 344±346. 10. Koike, H., Mamada, K., Ikeuchi, M. & Inoue, Y. (1989) Low molecular mass proteins in cyanobacterial photosystem II: iden- ti®cation of psbH and psbK gene products by N-terminal sequencing. FEBS Lett. 244, 3 91±396. 11. Abdel-Mawgood, A.L. & Dilley, R.A. (1990) Cloning and nucleotide sequence o f the psbH gene from cy anobacterium Synechocystis 6803 . Plant Mol. Biol. 14, 445±446. 12. Mayes, S.R. & Barber, J. (1990) Nucleotide sequence of the psbH gene of the cyanobacterium Synechocystis 6803. Nucleic Acid Res. 18, 194. 13. Komenda, J. & B arber, J. (1995) C omparison of psbO and psbH deletion mutants of Synechocystis PCC 68 03 indicate that degra- dation of D1 protein is regulated by the Q B site and is dependent on protein s ynthesis. Biochemist ry 34, 9 625±9631. 14. Williams, J.G.K. (1988) Construction of speci®c mutations in photosystem II photosynthetic reaction center by genetic engi- neering. Methods E nzymol. 167, 7 66±778. 15. Pakrasi, H.B., Williams, J .G.K. & Arntzen, C.J. (1 988) Targeted mutagenesis of the psb E and psbF genes blocks photosynthetic electron transport: evidence for a fu nctional role of cytochrome b559 in photosystem II. EMB O J. 7, 325±332. 16. Tang, X S . & D iner, B.A. ( 1994) Biochemical a nd spectroscopic characterization of a new oxygen-evolving ph otosystem II core complex from the cyanobacterium Synechocystis PCC 6803. Bio- chemistry 33, 4594±4603. 17. Ritter, S., Komenda, J., S Ï etlõ  kova  ,E.,S Ï etlõ  k, I. & Welte, W. (1992) Immobilized metal a nity chromatography for the separation o f photosystems I and II from the thermophilic cyanobacterium Synechococcus elongatus. J. Chr omatogr. 625, 21±31. 18. Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680±685. 19. Bartos Ï ,J.,Berkova  ,E.&S Ï etlõ  k, I. (1975) A versatile chamber for gas exchange measurements in suspensions of algae and chloro- plasts. Photosynthetica 9, 395±406. 20. Wellburn, A.R. (1984) The spectral determination of chlorophyll- a and chlorophyll-b, as well as carotenoids, using various solvents with spectrophotometers of dierent resolution. J. Plant. Physiol. 144, 307±313. 21. Gilmore, M.A. & Yamamoto, H.Y. (1991) Resolution of lutein and zeaxanthin using non-endcapped, lightly carbon-loaded C 18 high-performance liquid chromatograph ic column. J. Chromatogr. 543, 137±145. 22. Komenda, J. (2000) Role of two forms of the D1 protein in the recovery from photoinhibition of pho tosystem II in the cyano- bacterium Synechococcus PCC 7942. Biochim. Biophys. Acta 1457, 243±252. 23. Komenda, J. & Masojõ  dek, J. (1995) Oxygen retards recovery from phetoinhibition due to a secondary damage to the PSII complex. In Photosynthesis: From Light to Biosphere (Mathis, P. ed) Vol. IV, p p. 203±206. Kluwer Academic Press, The N etherlands. 24. Barbato,R.,Friso,G.,Rigoni,F.,Frizzo,A.&Giacometti,G.M. (1992) Characterization of a 41 kDa photoinhibition adduct in isolated photosystem II reaction centres. FEBS Lett. 309, 165±169. 25. Miyao,M.,Ikeuchi,M.,Yamamoto,N.&Ono,T.(1995)Speci®c degradation of the D1 protein of photosystem II by treatment with hydrogen peroxide in darkness: implications for the mechanism of degradation of the D1 protein under illumination. Biochemistry 34, 10019±10026. 26. He, W.Z., Newell, W .R., Haris, P.I., Chapman, D. & Barber, J. (1991) Protein secondary structure o f the isolated photosystem-II reaction center and conf ormational ch anges studied by Fou rier- transform infrared-spectroscopy. Biochemistry 30, 4 552±4559. 27. Dedner, N., Meyer, H.E ., Ashton , Ch & Wildner, G.F. (1988) N-terminal sequence analysis of the 8 kDa protein in Chlamydo- monas reinhardii. Localization of the phosphothreonine . FEBS Lett. 236, 77±82. 28. Hamel, P., Olive, J., Pierre, Y., Wollman, F A. & De Vitry, C. (2000) A new subunit of cytochrome b6f complex undergoes reversible phosphorylation up on state transition. J. Biol. Chem. 275, 17072±17079. 29. Race, H.L. & Gounaris, K. (1993) Identi®cation of the psbH gene- product as a 6-kDa phosphop rotein in the cyanobac terium Synechocystis 680 3. FEBS Lett. 323, 3 5±39. 30. Michel, H.P., Hunt, D.F., Shabanowitz, J. & Bennett, J. (1988) Tandem mass spectrometry reveals that three photosystem II proteins of spinach chloroplasts contain N-acetyl-O-pho spho- threonine a t their NH 2 termini. J. Biol. Chem. 263, 1123±1130. 31. Vener, A.V., Harms, A ., Sussman, M.R. & Vi erstra, R.D. (2001) Mass spectrometric r esolution of reversible protein phosphoryla- tion in photosynthetic membranes of Arabidopsis thaliana. J. Biol. Chem. 276, 6959±6966. 32. Rintama È ki, E., Salonen, M., Suoranta, U.M., Carlberg, I., Andersson, B. & Aro, E.M. (1997) Phosphorylation of light-har- vesting complex II and photosystem II core proteins shows dierent irradiance-d ependent regulation in vivo ± application of phosphothreonin e antib odies to analysis of thylakoid phospho- proteins. J. Biol. Chem. 272, 30476±30482. 33. Packham, N.K. (1988) Is the 9 kDa thylakoid membrane phos- phoprotein functionally and structurally analogous to the H subunit of bacterial reaction centres? FEBS Lett. 231, 2 84±290. 34. Chiaramonte, S., Giacometti, G.M. & Bergantino, E. (1999) Construction and characterization of a functional mutant of Synechocystis 6803 harbouring an eukaryotic PSII-H subun it. Eur. J. Biochem. 260, 833±843. 35. Zouni, A., Witt, H.T., Kern, J., Fromme, P., Krauss, N., Saenger, W. & Orth, P. (2001) Crystal structure of photosystem II from Synechococcus elongatus at 3.8 A Ê resolution. Nature 409, 739±743. 36. Buchel, C., Morris, E., Orlova, E. & Barber, J. (2001) Localisation of the PsbH subunit i n photosystem II: a new approach using labelling of His-tags with a Ni 2+ -NTA gold cluster and single particle analysis. J. Mol. Biol. 312, 371±379. 37. Zheleva, D., Sharma, J., Panico, M., Morris, H.R. & Barber, J. (1998) Isolation and characterization of monomeric and d imeric CP47- reaction center photosystem II complexes. J. Biol. Chem. 273, 16122±16127. 618 J. Komenda et al. (Eur. J. Biochem. 269) Ó FEBS 2002 38. Zhang, L.X., Paakkarinen, V., Vanwijk, K.J. & Aro, E.M. (1999) Co-translational assembly of the D1 protein into photosystem II. J. Biol. C hem. 274, 16062±16067. 39. Constant, S., Perewoska, I., Alfonso, M. & Kirilovsky, D. (1997) Expression of the psbA gene during photoinhibition and recovery in Synechocystis PCC 6714: inhibition and damage of transcrip- tional and translational machinery prevent the restoration of photosystem II activity. Plant M ol. Biol. 34, 1 ±13. 40. Giardi, M.T., Komenda, J. & Mas ojidek, J. (1994) Involvement of protein phosphorylation in the sensitivity of photosystem II to strong illumination. Physiol. Plantarum 92, 181±187. 41. Xiong, J ., Subramaniam, S. & Govindjee (1996 ) Modeling of t he D1/D2 proteins and cofactors of the photosystem II reaction center: implications for herbicide and bicarbonate binding. Protein Sci. 5, 2054±2073. 42. Sundby, C., Ulla, K., Henrysson, L. & Henrysson, T. (1989) Eects of bicarbonate on thylakoid protein phosphorylation. Biochim. Biop hys. Acta 975 , 277±282. 43. Nixon, P.J., Komenda, J., Barber, J., Deak, S., Vass, I. & D iner, B.A. (1995) Deletion of the PEST-like region of p hotosystem II modi®es the QB-binding pocket but does not p revent rapid turnover of D1. J. Biol. C hem. 270, 14919±14927. 44. Graan, T. & Ort, D.R. (1986) Detection of oxygen evolving photosystem II centers in active in plastoquinone redu ction. Bio- chim. Biophys. A cta 852, 320±330. Ó FEBS 2002 Function of the PsbH protein in photosystem II (Eur. J. Biochem. 269) 619 . Absence of the psbH gene product destabilizes photosystem II complex and bicarbonate binding on its acceptor side in Synechocystis PCC 6803 Josef. y i nhibition of the D1± D2 adduct formation in the I C7 strain. The PsbH protein affects the bicarbonate binding on the acceptor side of PSII A characteristic