Oxidative stress inhibits the repair of photodamage to the photosynthetic machinery
2001; Springer Nature; Volume: 20; Issue: 20 Linguagem: Inglês
10.1093/emboj/20.20.5587
ISSN1460-2075
Autores Tópico(s)Photoreceptor and optogenetics research
ResumoArticle15 October 2001free access Oxidative stress inhibits the repair of photodamage to the photosynthetic machinery Yoshitaka Nishiyama Yoshitaka Nishiyama Department of Regulation Biology, National Institute for Basic Biology, Myodaiji, Okazaki, 444-8585 Japan Department of Biomechanics, School of Life Science, The Graduate University for Advanced Studies, Myodaiji, Okazaki, 444-8585 Japan Search for more papers by this author Hiroshi Yamamoto Hiroshi Yamamoto Plant Molecular Physiology Laboratory, Research Institute of Innovative Technology for the Earth (RITE), Kizu, Kyoto, 619-0292 Japan Search for more papers by this author Suleyman I. Allakhverdiev Suleyman I. Allakhverdiev Department of Regulation Biology, National Institute for Basic Biology, Myodaiji, Okazaki, 444-8585 Japan Search for more papers by this author Masami Inaba Masami Inaba Department of Regulation Biology, National Institute for Basic Biology, Myodaiji, Okazaki, 444-8585 Japan Search for more papers by this author Akiho Yokota Akiho Yokota Plant Molecular Physiology Laboratory, Research Institute of Innovative Technology for the Earth (RITE), Kizu, Kyoto, 619-0292 Japan Graduate School of Biological Sciences, Nara Institute of Science and Technology (NAIST), Ikoma, Nara, 630-0101 Japan Search for more papers by this author Norio Murata Corresponding Author Norio Murata Department of Regulation Biology, National Institute for Basic Biology, Myodaiji, Okazaki, 444-8585 Japan Department of Biomechanics, School of Life Science, The Graduate University for Advanced Studies, Myodaiji, Okazaki, 444-8585 Japan Search for more papers by this author Yoshitaka Nishiyama Yoshitaka Nishiyama Department of Regulation Biology, National Institute for Basic Biology, Myodaiji, Okazaki, 444-8585 Japan Department of Biomechanics, School of Life Science, The Graduate University for Advanced Studies, Myodaiji, Okazaki, 444-8585 Japan Search for more papers by this author Hiroshi Yamamoto Hiroshi Yamamoto Plant Molecular Physiology Laboratory, Research Institute of Innovative Technology for the Earth (RITE), Kizu, Kyoto, 619-0292 Japan Search for more papers by this author Suleyman I. Allakhverdiev Suleyman I. Allakhverdiev Department of Regulation Biology, National Institute for Basic Biology, Myodaiji, Okazaki, 444-8585 Japan Search for more papers by this author Masami Inaba Masami Inaba Department of Regulation Biology, National Institute for Basic Biology, Myodaiji, Okazaki, 444-8585 Japan Search for more papers by this author Akiho Yokota Akiho Yokota Plant Molecular Physiology Laboratory, Research Institute of Innovative Technology for the Earth (RITE), Kizu, Kyoto, 619-0292 Japan Graduate School of Biological Sciences, Nara Institute of Science and Technology (NAIST), Ikoma, Nara, 630-0101 Japan Search for more papers by this author Norio Murata Corresponding Author Norio Murata Department of Regulation Biology, National Institute for Basic Biology, Myodaiji, Okazaki, 444-8585 Japan Department of Biomechanics, School of Life Science, The Graduate University for Advanced Studies, Myodaiji, Okazaki, 444-8585 Japan Search for more papers by this author Author Information Yoshitaka Nishiyama1,2, Hiroshi Yamamoto3, Suleyman I. Allakhverdiev1, Masami Inaba1, Akiho Yokota3,4 and Norio Murata 1,2 1Department of Regulation Biology, National Institute for Basic Biology, Myodaiji, Okazaki, 444-8585 Japan 2Department of Biomechanics, School of Life Science, The Graduate University for Advanced Studies, Myodaiji, Okazaki, 444-8585 Japan 3Plant Molecular Physiology Laboratory, Research Institute of Innovative Technology for the Earth (RITE), Kizu, Kyoto, 619-0292 Japan 4Graduate School of Biological Sciences, Nara Institute of Science and Technology (NAIST), Ikoma, Nara, 630-0101 Japan *Corresponding author. E-mail: [email protected] The EMBO Journal (2001)20:5587-5594https://doi.org/10.1093/emboj/20.20.5587 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Absorption of excess light energy by the photosynthetic machinery results in the generation of reactive oxygen species (ROS), such as H2O2. We investigated the effects in vivo of ROS to clarify the nature of the damage caused by such excess light energy to the photosynthetic machinery in the cyanobacterium Synechocystis sp. PCC 6803. Treatments of cyanobacterial cells that supposedly increased intracellular concentrations of ROS apparently stimulated the photodamage to photosystem II by inhibiting the repair of the damage to photosystem II and not by accelerating the photodamage directly. This conclusion was confirmed by the effects of the mutation of genes for H2O2-scavenging enzymes on the recovery of photosystem II. Pulse labeling experiments revealed that ROS inhibited the synthesis of proteins de novo. In particular, ROS inhibited synthesis of the D1 protein, a component of the reaction center of photosystem II. Northern and western blot analyses suggested that ROS might influence the outcome of photodamage primarily via inhibition of translation of the psbA gene, which encodes the precursor to D1 protein. Introduction Oxygen is essential for most living organisms but is also a precursor to reactive oxygen species (ROS), which can damage cellular components such as proteins, lipids and nucleic acids (Halliwell and Gutteridge, 1990). In oxygenic photosynthesis, various ROS, such as the superoxide radical (O2−), hydrogen peroxide (H2O2) and the hydroxyl radical (·OH), are generated as a result of the photosynthetic transport of electrons (Asada, 1996, 1999). When the absorption of light energy by chlorophylls exceeds the capacity for utilization of the energy in photo synthesis, the generation of ROS is greatly accelerated. High-intensity light is an environmental factor that can damage the photosynthetic machinery of plants and it is often the cause of reductions in the growth and productivity of crop plants (Long et al., 1994). The main target of photodamage is photosystem II (PSII), which is a complex of proteins and pigments that is the site of the photochemical reaction and the subsequent transport of electrons from water to plastoquinone. Photodamage to PSII is mainly due to damage to the D1 protein, which forms a heterodimer with the D2 protein in the reaction center of PSII, and the subsequent rapid degradation of the D1 protein (Barber and Andersson, 1992; Prásil et al., 1992; Aro et al., 1993; Andersson and Barber, 1996). A number of reports have suggested that ROS are involved in the photodamage to PSII (Prásil et al., 1992; Aro et al., 1993; Andersson and Barber, 1996). ROS such as O2− (Chen et al., 1992), H2O2 (Ananyev et al., 1992; Miyao et al., 1995) and ·OH (Miyao et al., 1995) were shown to induce the cleavage of the D1 protein directly in vitro. Besides such ROS, 1O2, which is generated by transfer of excitation energy in the reaction center of PSII, is often considered to trigger degradation of the D1 protein (Vass et al., 1992; Hideg et al., 1994; Telfer et al., 1994; Okada et al., 1996). Since the cited studies were conducted in vitro using thylakoid membranes or PSII complexes, the action of ROS in living cells in which the ROS-scavenging system is fully active is still in question. In the present study, we examined whether ROS, such as H2O2, act primarily by accelerating the damage to PSII directly or by inhibiting the repair of the photodamage to PSII in vivo, using the cyanobacterium Synechocystis sp. PCC 6803 (hereafter Synechocystis). We found that increased intracellular concentrations of ROS inhibited repair of the photodamage to PSII. Labeling of proteins in vivo and northern and western blot analyses demonstrated that ROS primarily inhibited the synthesis of the D1 protein de novo at the translational level at both the initiation and elongation steps. Results Light and ROS act synergistically in the photodamage to PSII Incubation of Synechocystis cells in strong light at 1.5 mmol photons/m2/s had no significant effect on the activity of PSII over a 2-h period (Figure 1A; open circles). However, in the presence of chloramphenicol, an inhibitor of protein synthesis, similar incubation resulted in the inactivation of PSII (Figure 1B; open circles). These observations suggested that, in light at 1.5 mmol photons/m2/s, the rate of repair of photodamage that required the synthesis of proteins de novo was higher than the rate of photodamage to PSII. Figure 1.Light-induced inactivation of PSII in the presence of reagents that accelerate the production of ROS in wild-type Synechocystis. Cells were exposed to light at 1.5 mmol photons/m2/s with standard aeration. (A) In the presence of methyl viologen at 2 μM (open triangles) and at 5 μM (open squares) and in its absence (open circles); filled circles: in the presence of 5 μM methyl viologen in darkness. (B) The same as (A) but in the presence of 200 μg/ml chloramphenicol. (C) In the presence of H2O2 at 0.5 mM (open triangles) and at 2 mM (open squares) and in its absence (open circles); filled circles: in the presence of 2 mM H2O2 in darkness. (D) The same as (C) but in the presence of 200 μg/ml chloramphenicol. PSII activity was monitored in terms of the photosynthetic evolution of oxygen in the presence of 1 mM 1,4-benzoquinone as the electron acceptor. The activity taken as 100% was 527 ± 46 μmol O2/mg chlorophyll/h. Values are means ± SD (bars) of results from four independent experiments. Download figure Download PowerPoint We next examined the actions of ROS by modulating their intracellular concentrations in cells using reagents that induce the production of ROS. Methyl viologen is an acceptor of electrons that accelerates the production of O2− and H2O2 in light (Asada, 1996). The presence of methyl viologen during the incubation of cells in light at 1.5 mmol photons/m2/s enhanced the apparent photodamage to PSII (Figure 1A; open triangles and squares). However, when repair of PSII was blocked by the presence of chloramphenicol, the effect of methyl viologen was abolished (Figure 1B; open triangles and squares). In darkness, methyl viologen had no effect on the activity of PSII (Figure 1A; closed circles). Exogenously supplied H2O2 also accelerated the photodamage to PSII (Figure 1C), suggesting that the agent that accelerated photodamage during incubation in the presence of methyl viologen might have been H2O2. The presence of chloramphenicol during the incubation further accelerated the photodamage and abolished the effect of H2O2 (Figure 1D). Incubation in darkness in the presence of H2O2 had no effect on the activity of PSII (Figure 1C; closed circles). These observations suggested that the enhanced photodamage to PSII in living cells might have been due to inhibition of the repair of PSII and not to direct damage to PSII. A mutant defective in H2O2-scavenging enzymes with enhanced sensitivity to photodamage to PSII Enzymes that scavenge H2O2, such as catalase peroxidase (Tichy and Vermaas, 1999) and thioredoxin peroxidase (Yamamoto et al., 1999), which are encoded by the katG and tpx genes, respectively, appear to maintain intracellular concentrations of H2O2 at low and non-toxic levels. Targeted mutagenesis of both the katG and tpx genes in Synechocystis cells enhanced photodamage to PSII but had no significant effect on such photodamage in the presence of chloramphenicol (Figure 2). These observations supported the hypothesis that ROS might inhibit the repair of photodamage to PSII. Figure 2.Light-induced inactivation of PSII in wild-type Synechocystis and in katG−/tpx− mutant cells. Cells were exposed to light at 1.5 mmol photons/m2/s in the presence of 200 μg/ml chloramphenicol and in its absence. The other experimental conditions were the same as described in the legend to Figure 1. (A) Wild-type cells. (B) katG−/tpx− cells. Open triangles and circles: in the presence and absence of chloram phenicol, respectively. PSII activity was monitored in terms of the photosynthetic evolution of oxygen. The oxygen-evolving activity of katG−/tpx− cells that was taken as 100% was 560 ± 44 μmol O2/mg chlorophyll/h. Values are means ± SD (bars) of results from four independent experiments. Download figure Download PowerPoint Inhibition by exogenous H2O2 of the repair of photodamage to PSII We examined inhibition of repair by H2O2 directly by monitoring the recovery of PSII activity after photodamage (Figure 3). When wild-type cells were incubated in strong light at 3.0 mmol photons/m2/s, PSII activity fell to 20% of the original level. During subsequent exposure of cells to weak light, the activity of PSII returned to the original level (Figure 3A). The presence of H2O2 during the incubation in weak light suppressed the repair of the photodamage to PSII. Targeted mutagenesis of the katG and tpx genes delayed the repair of photodamage (Figure 3B). Furthermore, repair in katG−/tpx− mutant cells was completely eliminated in the presence of 0.5 mM H2O2. Removal of H2O2 from the mutant cells by exogenously supplied catalase allowed repair to proceed (Figure 3B; open diamonds). Figure 3.Repair of photodamage to PSII and the effects of H2O2 after light-induced inactivation in wild-type Synechocystis and katG−/tpx− cells. Cells were exposed to light at 3 mmol photons/m2/s for 60 min (wild type) or 40 min (katG−/tpx−) without aeration to induce ∼80% inactivation of PSII. Cells were then incubated in light at 70 μmol photons/m2/s with standard aeration in the presence of 0.5 mM H2O2 (open triangles), 2 mM H2O2 (open squares) or 200 μg/ml chloramphenicol (filled triangles), and in the absence of these reagents (open circles). (A) Wild-type cells. (B) katG−/tpx− cells. Diamond symbols indicate repair after 0.1 μM catalase had been added, at the time indicated by the vertical arrow, to a suspension of cells that had been incubated in the presence of 0.5 mM H2O2. Values are means ± SD (bars) of results from three independent experiments. Download figure Download PowerPoint Inhibition by ROS of the synthesis of the D1 protein We used western blot analysis to examine changes in the level of the D1 protein when intracellular concentrations of ROS were increased by the presence of methyl viologen during incubation in light at 1.5 mmol photons/m2/s (Figure 4). The presence of methyl viologen decreased the apparent level of D1 protein. However, the decrease in the level of pre-existing D1 protein, assessed in the presence of chloramphenicol, was unaffected by methyl viologen. Therefore, we postulated that ROS, produced by the photosynthetic transport of electrons that was mediated by methyl viologen, inhibited the synthesis of the D1 protein de novo but did not affect the degradation of this protein. Figure 4.Changes in the level of the D1 protein during the light-induced inactivation of PSII in wild-type cells. (A) Results of western blot analysis. (B) Quantitation of the results shown in (A). Cells were exposed to light at 1.5 mmol photons/m2/s with standard aeration in the absence (lane 1; open circles) and in the presence (lane 2; open triangles) of 5 μM methyl viologen. Cells were also exposed to light in the presence of 200 μg/ml chloramphenicol (lane 3; filled circles) and in the presence of 200 μg/ml chloramphenicol plus 5 μM methyl viologen (lane 4; filled triangles). Thylakoid membranes were isolated from cells at the indicated times. Values are means ± SD (bars) of results from three independent experiments. Download figure Download PowerPoint We examined the above-mentioned hypothesis by monitoring the incorporation of [35S]methionine into proteins and, in particular, into the D1 protein (Figure 5). The synthesis of the D1 protein de novo was markedly suppressed in the presence of methyl viologen and H2O2. The rate of synthesis of the D1 protein in katG−/tpx− cells was lower than that in wild-type cells, suggesting that inhibition by ROS of the synthesis of the D1 protein in katG−/tpx− cells might have been more severe than that in wild-type cells. Furthermore, not only the synthesis of the D1 protein de novo but also the synthesis of almost all other proteins was suppressed in the presence of ROS (Figure 5A). Figure 5.The synthesis of the D1 protein de novo in the presence of ROS in wild-type Synechocystis and in katG−/tpx− cells, as monitored in terms of the incorporation of radioactive [35S]methionine into proteins of thylakoid membranes. (A) Lanes 1–3, wild-type cells were labeled for 20 min in the absence of added reagents (lane 1), in the presence of 5 μM methyl viologen (lane 2) and in the presence of 2 mM H2O2 (lane 3). Lane 4, katG−/tpx− cells were labeled in the absence of any reagent. The arrow indicates the D1 protein. (B) Changes in the level of labeled D1 protein that had been incorporated into thylakoid membranes in wild-type cells in the absence of any added reagents (open circles), in the presence of 5 μM methyl viologen (open triangles), and in the presence of 0.5 mM H2O2 (open squares); filled circles indicate labeled D1 protein in katG−/tpx− cells in the absence of any added reagents. Values are means ± SD (bars) of results from three independent experiments. Download figure Download PowerPoint Effects of ROS on the expression of psbA genes To determine whether ROS might inhibit the synthesis of the D1 protein at the transcriptional level, we examined the expression of psbA genes for the D1 protein in the presence of increased intracellular concentrations of ROS. In Synechocystis, the D1 protein is encoded by a small multigene family that consists of the psbA1, psbA2 and psbA3 genes (Jansson et al., 1987). The psbA1 gene is non-functional, while the psbA2 and psbA3 genes are expressed in response to light (Tyystjärvi et al., 1998) and both encode the identical D1 protein. We monitored changes in the level of a mixture of psbA2 and psbA3 transcripts by northern blot analysis (Figure 6). The presence of methyl viologen did not affect the light-induced expression of the psbA genes. At 0.5 mM, H2O2 did not significantly affect the induction of the expression of these genes at the early stage (zero to 10 min) of incubation. However, at 20 min it repressed the level of psbA transcripts to 70% of the control level. Targeted mutagenesis of the katG and tpx genes had no effect on induction of the expression of the psbA genes. Inhibition by chloramphenicol of the translation of the psbA transcripts also had no effect on the induction of expression during a 20-min incubation, and the level of psbA transcripts was maintained subsequently. These observations suggest that the inhibition by ROS during the synthesis of the D1 protein de novo might not occur primarily at transcription. Figure 6.Effects of ROS on the light-induced expression of psbA genes in wild-type Synechocystis and katG−/tpx− cells. (A) Results of northern blot analysis. (B) Quantitation of the results in (A). Prior to exposure to light, cells were incubated at 30°C in darkness for 60 min. Wild-type cells were exposed to light at 1.5 mmol photons/m2/s in the absence of any added reagents (control; open circles) and in the presence of 5 μM methyl viologen (MV; open triangles), 0.5 mM H2O2 (open squares) and 200 μg/ml chloramphenicol (Cm; filled triangles); filled circles indicate induction of the expression of psbA genes in katG−/tpx− cells in the absence of any added reagents. Total RNA (5 μg) was loaded in each lane of the gel. Values are means ± SD (bars) of results from three independent experiments. Download figure Download PowerPoint We also examined the action of ROS on the synthesis of the precursor to D1 protein (pre-D1) at the translational level. Figure 7 shows that the level of pre-D1 increased during exposure of wild-type cells to strong light and reached a maximum at 40 min. The presence of 0.5 mM H2O2 significantly depressed induction of the synthesis of pre-D1. These observations suggest that the main target of inhibition by ROS during the synthesis of the D1 protein might be translation. Figure 7.Changes in the level of the pre-D1 protein during the light-induced inactivation of PSII in wild-type cells. Cells were exposed to light at 1.5 mmol photons/m2/s in the absence of H2O2 (open circles), in the presence of 0.5 mM H2O2 (open squares), and in the presence of 200 μg/ml chloramphenicol (open triangles). Thylakoid membranes were isolated from cells at the indicated times. Values are means ± SD (bars) of results from three independent experiments. Download figure Download PowerPoint Effects of ROS on the translation of psbA mRNAs To identify the step in the translation of psbA mRNAs that was inhibited by ROS, we examined the effects of H2O2 on the distribution of psbA mRNAs in membrane-bound and cytosolic polysomes as well as in a polysome-free form in wild-type cells (Figure 8). In Synechocystis, psbA mRNAs are first associated with polysomes in the cytosol (translation initiation) and, after translational elongation has proceeded to a certain extent, psbA mRNA–ribosome complexes are targeted to thylakoid membranes leading to the subsequent elongation (Tyystjärvi et al., 2001; see Discussion). Figure 8.Effects of H2O2 on the distribution of psbA mRNAs that are free or are associated with polysomes in wild-type cells. Cells were incubated at 30°C in darkness for 60 min and then exposed to light at 1.5 mmol photons/m2/s for 20 min in the presence of 0.5 mM H2O2 and in its absence. Cells were disrupted and membrane-bound polysomes (M), cytosolic polysomes (C) and polysome-free RNA (F) were prepared as described in the text. RNA was isolated from each fraction and subjected to northern blotting with a labeled fragment of the psbA2 gene as the probe. The amount of RNA in each lane corresponded to that from each individual fraction, and each fraction was derived from cells equivalent to 5 μg chlorophyll. (A) Gel-electrophoretic pattern. The results shown are representative of the results of four independent experiments, each of which gave similar results. (B) Quantified results. Values are means ± SD (bars) of results from four independent experiments. T represents the total of the three fractions of psbA mRNA. Download figure Download PowerPoint Figure 8 shows that, during incubation in light at 1.5 mmol photons/m2/s, almost all the psbA mRNA was in the form of polysomes, which were either associated with or dissociated from thylakoid membranes. The level of psbA mRNA that was not associated with polysomes was very low. The presence of 0.5 mM H2O2 significantly decreased the relative level of psbA mRNAs in membrane-associated polysomes. These findings suggest that the elongation step in the translation of psbA mRNAs might be strongly inhibited by H2O2. In addition, the presence of 0.5 mM H2O2 increased the level of psbA mRNA that was not associated with polysomes, suggesting that the binding of psbA mRNAs to ribosomes, namely the initiation of translation, might be also inhibited by H2O2. Discussion In the present study we investigated the actions of ROS in the photodamage to PSII and its repair in Synechocystis cells. Conditions that produced ROS stimulated the apparent photodamage to PSII (Figure 1). Thus, it appeared that ROS might act primarily by inhibiting repair of the photodamage to PSII in vivo (Figure 3A). Inactivation of H2O2-scavenging enzymes also inhibited the repair of the photodamage to PSII (Figure 3B). Which ROS affect the repair of photodamage? When the light energy absorbed by the photosynthetic machinery exceeds the energy that is required for photosynthesis, O2− is generated on the acceptor side of photosystem I (Asada, 1996). In Synechocystis, O2− is disproportionated by an Fe-containing superoxide dismutase to H2O2, which is then scavenged by catalase peroxidase and thioredoxin peroxidase (Tichy and Vermaas, 1999; Yamamoto et al., 1999). However, strongly oxidative conditions, for example, the combination of light and the presence of methyl viologen, allow the photosynthetic machinery to produce excessive amounts of O2− and H2O2. In such cases, a small fraction of the H2O2 is converted to the hydroxyl radical ·OH by the action of transition metal cations, such as Fe2+ and Cu2+ (Asada, 1996). Among the various ROS, O2− is highly reactive and ·OH is extremely reactive, whereas H2O2, which is the most abundant ROS, is the least reactive. We can ask which of the three ROS is active in the inhibition of repair. It is unlikely that O2− is the major contributor to the inhibition because exogenously supplied H2O2, which cannot be converted to O2−, was as effective as the presence of methyl viologen. Inhibition by ROS of the repair of photodamage to PSII was reversible (Figure 3B), an observation that suggests that the repair system might be inhibited directly by H2O2 and/or ·OH and not by the secondary products of ROS, such as lipid peroxides (Wise, 1995). Effects of ROS on the de novo synthesis of proteins Western blot analysis indicated that ROS did not enhance the degradation of the pre-existing D1 protein when protein synthesis was blocked by the presence of chloramphenicol. However, degradation of the pre-existing D1 protein occurred more slowly than inactivation of PSII. This difference might have been caused by the presence of inactivated D1 protein, which was detected by western blot analysis. Since degradation of inactivated D1 protein is coordinated with the synthesis de novo of active D1 protein (Komenda and Barber, 1995), it is likely that inhibition of its synthesis de novo by either ROS or chloramphenicol led to the accumulation of inactivated D1 protein. Labeling of proteins in vivo provided direct evidence for the inhibition by ROS of the synthesis of the D1 protein de novo (Figure 5). Thus, the enhanced sensitivity of the katG−/tpx− cells to strong light might be explained by the inhibition by ROS of the synthesis of the D1 protein de novo. It is noteworthy, in this context, that not only the synthesis of the D1 protein de novo but also the synthesis of almost all other proteins was suppressed in the presence of ROS (Figure 5). Thus, it seems likely that ROS might inhibit the synthesis of most proteins in Synechocystis. Translational initiation and elongation as the primary target of ROS Northern blot analysis indicated that ROS had only a slight effect on the accumulation of psbA transcripts (Figure 6). Thus, since ROS inhibited the synthesis of pre-D1 de novo (Figures 5 and 7), it seemed likely that ROS might act primarily by inhibiting translation. The distribution of psbA mRNAs (Figure 8) revealed that ROS markedly decreased the level of psbA mRNA–polysome complexes that were associated with thylakoid membranes. Recently, Tyystjärvi et al. (2001) demonstrated that, in Synechocystis, synthesis of the D1 protein de novo is regulated at the level of transcription of psbA genes and at the elongation step in the translation of psbA mRNA. The transcriptional machinery in Synecho cystis is different from that in chloroplasts. In the latter, the synthesis of D1 protein is redox-regulated at the initiation of the translation of psbA mRNA via binding of the nuclear-encoded protein complex to the 5′-untranslated region of psbA mRNA (Hirose and Sugiura, 1996; Yohn et al., 1996; Trebitsh et al., 2000) In Synechocystis, psbA mRNAs associate first with ribosomes in the cytosol (translation initiation) and then elongation starts. After a 17-kDa nascent chain of the D1 protein has been synthesized, the psbA mRNA–ribosome complexes are targeted to thylakoid membranes where further elongation of the D1 protein occurs (Tyystjärvi et al., 2001). Our observations together with these observations suggest that the major sites of inhibition by ROS in our system are both initiation and elongation steps of translation. However, it remains possible that ROS might also inhibit the targeting of psbA mRNA–polysome complexes to thylakoid membranes. Ayala et al. (1996) demonstrated that oxidative stress due to cumene hydroperoxide inhibits protein synthesis in rat liver by inactivating elongation factor 2, an important participant in the elongation step of translation. In Escherichia coli, H2O2 inactivates elongation via carbonylation of elongation factor G (Tamarit et al., 1998). Our results are consistent with these earlier findings. Materials and methods Cells and culture conditions Wild-type Synechocystis sp. PCC 6803 and katG−/tpx− mutant cells were grown photoautotrophically (Gombos et al., 1994) at 34°C in BG-11 medium under light at 70 μmol photons/m2/s, with aeration by sterile air that contained 1% CO2. Cells at a density of 5 ± 0.5 μg/ml chlorophyll were used directly for studies of photodamage and repair. Conditions for photodamage and repair, and measurements of photosynthetic activity Cells were incubated in light at 1.5 mmol photons/m2/s (unless otherwise noted) to induce photodamage to PSII or in light at 70 μmol photons/m2/s to induce the repair of photodamaged PSII. These experiments were performed at 30°C at a concentration of cells that corresponded to 5 ± 0.5 μg/ml chlorophyll. The activity of PSII was measured at 30°C by monitoring the photosynthetic evolution of oxygen, which was determined from the concentration of oxygen in a suspension of cells, measured with a Clark-type oxygen electrode in the presence of 1.0 mM 1,4-benzoquinone, as described previously (Gombos et al., 1994). Concentrations of chlorophyll were determined as described by Arnon et
Referência(s)