Herbicide Effect on the Photodamage Process of Photosystem II: Fourier Transform Infrared Study

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

Biochimica et Biophysica Acta 1807 (2011) 1214–1220

Contents lists available at ScienceDirect

Biochimica et Biophysica Acta

journal homepage: www.elsevier.com/locate/bbabio

Herbicide effect on the photodamage process of photosystem II: Fourier transform infrared study

Issei Idedan b, Tatsuya Tomo c,d, Takumi Noguchi a,b,⁎

a Division of Material Science, Graduate School of Science, Nagoya University, Furo-cho, Chikusa-ku, Nogoya, 464-8602, Japan b Institute of Materials Science, University of Tsukuba, Tsukuba, Ibaraki, 305-8573, Japan c Department of Biology, Faculty of Science, Tokyo University of Science, Kagurazaka 1-3, Shinjuku-ku, Tokyo, 162-8601, Japan d PRESTO, Japan Science and Technology Agency (JST), 4-1-8 Honcho Kawaguchi, Saitama, 332-0012, Japan

article info abstract

Article history: The photodamage process of photosystem II by strong illumination was investigated by examining the Received 14 May 2011 herbicide effects on the photoinactivation of redox cofactors. O2-evolving photosystem II membranes from Received in revised form 7 June 2011 spinach in the absence of herbicide and in the presence of 3-(3,4-dichlorophenyl)-1,1-dimethylurea (DCMU) Accepted 7 June 2011 and bromoxynil were subjected to strong white-light illumination at 298 K, and the illumination-time Available online 28 June 2011 dependence of the activities of QA, the Mn cluster, and P680 were monitored using light-induced Fourier transform infrared (FTIR) difference spectroscopy. The decrease in the Q activity was suppressed and Keywords: A FTIR accelerated by DCMU and bromoxynil, respectively, in comparison with the sample without herbicide. The Herbicide intensity change in the S2/S1 FTIR signal of the Mn cluster exhibited a time course virtually identical to that in Photodamage the QA signal in all the three samples, suggesting that the loss of the S1 →S2 transition was ascribed to the QA

Photoinhibition inactivation and hence the Mn cluster was inactivated not faster than QA. The decrease in the P680 signal was Photoprotection always slower than that of QA keeping the tendency of the herbicide effect. Degradation of the D1 protein Photosystem II occurred after the P680 inactivation. These observations are consistent with the acceptor-side mechanism, in 1 which double reduction of QA triggers the formation of O2* to promote further damage to other cofactors and the D1 protein, rather than the recently proposed mechanism that inactivation of the Mn cluster initiates the photodamage. Thus, the results of the present study support the view that the acceptor-side mechanism dominantly occurs in the photodamage to PSII by strong white-light illumination. © 2011 Elsevier B.V. All rights reserved.

1. Introduction double reduction of QA leading to the release from the binding site as plastoquinol, PQH2. In the absence of QA, the triplet state formed on + − Plants utilize sunlight for their activities through light-energy ChlD1 by P680 Pheo charge recombination, has a relatively long 1 conversion processes, photosynthesis. However, light is also harmful lifetime (t1/2 ~1 ms) [10,11] and readily forms harmful O2*to to plants and the photosynthetic apparatus is damaged by light inactivate PSII. The triplet state is also formed by charge recombina- − − illumination, which is known as photoinhibition [1–8]. The major tion of S2,3QB or S2QA induced by low-light or flash illumination [12– target of photodamage is known to be photosystem II (PSII), where 14], which can thus be called the low-light mechanism. This electron transfer reactions are inactivated and the D1 polypeptide is mechanism may explain the reason why photodamage to PSII occurs specifically degraded. even under weak illumination. The donor-side mechanism has been Several different mechanisms of photodamage to PSII have been proposed for PSII in which the Mn cluster is already inactivated. In proposed so far [4,7]. In the acceptor-side mechanism [9], strong such PSII, oxidative damage to the electron donor side is caused by a − + illumination stabilizes QA semiquinone anion and then induces strong oxidant, P680 [15–17]. Recently, a new mechanism called the Mn mechanism [7,18] or the two-step mechanism [6,19] has been proposed. In this mechanism, the first step of photodamage is Abbreviations: ChlD1, monomeric chlorophyll on the D1 side of PSII; DCMU, 3-(3,4- dichlorophenyl)-1,1-dimethylurea; FTIR, Fourier transform infrared; Mes, 2-(N- destruction of the Mn cluster by its light absorption and then PSII is morpholino)ethanesulfonic acid; P680, special pair chlorophylls in PSII; Pheo, redox- inactivated via electron transfer reactions. This mechanism explains active pheophytin in PSII; PQH2, plastoquinol; PSII, photosystem II; QA, primary the previous observation that the quantum yield of photodamage is quinone electron acceptor; QB, secondary quinone electron acceptor not dependent on light intensity [20]. ⁎ Corresponding author at: Division of Material Science, Graduate School of Science, Thus, in spite of extensive studies, the molecular mechanism of Nagoya University, Furo-cho, Chikusa-ku, Nagoya, 464-8602, Japan. Tel.: +81 52 789 fi 2881; fax: +81 52 789 2883. photodamage to PSII has not been clari ed yet. In particular, the E-mail address: [email protected] (T. Noguchi). acceptor-side mechanism and the Mn mechanism provide conflicting

ideas about the initial target of photodamage, i.e., QA or the Mn (Oxford ITC-4). Single-beam spectra for 150 s (300 scans) were cluster, under strong illumination. Also, the detailed process of recorded before and after 5-s illumination by continuous white light inactivation of redox cofactors and the specific damage that triggers (~40 mW cm− 2) from a halogen lamp (Sigma Koki PHL-150), and a the D1 degradation have not been clearly resolved. light-minus-dark difference spectrum was calculated to represent the − − One of the clues to address these questions is herbicide effects on the S2QA /S1QA difference (or QA /QA and other donor-side components photodamage process. Herbicides bound to the QB pocket block the when the Mn cluster is inactivated before QA). The noise level was electron transfer beyond QA and also change the redox potential of QA estimated from a dark-minus-dark difference spectrum taken before differently depending on the herbicide species; a urea-type herbicide, illumination. − + DCMU, increases the QA /QA redox potential, whereas a phenol-type P680 /P680 FTIR measurement was performed as described herbicide, bromoxynil, decreases the potential [21]. Thus, herbicide previously [38] with slight modification. The PSII sample was treatment perturbs the acceptor-side photoreactions in PSII, which suspended in buffer A, and QA as well as the Mn cluster was depleted should affect the photodamage process. The herbicide effects on by treatment with 100 mM sodium dithionite and 30 μM benzyl photodamage studied so far [22–29] have shown a general tendency viologen followed by dark incubation for 5 h at room temperature that urea- and phenol-type herbicides retard and accelerate, respec- [39]. The sample was washed twice and resuspended in buffer A in the tively, the inactivation of electron transfer reactions and D1 degradation, presence of 40 mM potassium ferricyanide and 0.6 mM SiMo. After although some controversial results have been obtained in the effect of centrifugation at 170,000×g for 40 min, the pellet was sandwiched phenol-type herbicides on D1 degradation [22–24]. In such studies, between the BaF2 plates. The sample temperature was adjusted to − 2+ fluorescence [25,26,28],SiMoreduction[27], the QA Fe EPR signal 265 K. FTIR scans under dark (1 s) and during illumination (1 s) were 1 [25],and O2*production[29] have been detected to examine the repeated 500 times, and average single-beam spectra were used to photodamage to PSII. However, the inactivation processes of individual calculate a P680+/P680 difference spectrum. Light-illumination was redox cofactors in the presence of different-types of herbicides, which performed with red light (~16 mW cm2 at the sample surface) from a are necessary to determine the mechanism of photodamage, have not halogen lamp (Sigma Koki PHL-150) equipped with a red cutoff filter been systematically studied. (N600 nm). In this study, we have investigated the molecular mechanism of photodamage to O2-evolving PSII membranes under strong white- 2.3. Detection of D1 degradation light illumination by utilizing the herbicide effects on the inactivation processes of redox cofactors and the D1 protein degradation. For SDS-PAGE was performed using a gradient gel containing 16– detecting the activities of individual redox cofactors, we used light- 22% acrylamide and 7.5 M urea [40]. PSII samples equivalent to 1 μg induced Fourier transform (FTIR) difference spectroscopy, which is a of Chl/lane (before illumination) were loaded to the wells. For powerful method to directly monitor the photoreactions of redox immunological assays, the SDS–polyacrylamide gel was transferred cofactors in PSII [30–34]. We focused on the activities of QA, the Mn onto a polyvinylidene fluoride (PVDF) membrane, and subjected to cluster and P680, which are the key cofactors in the proposed the reaction with an antibody raised against the C-terminus of the photodamage mechanisms [4,7], and studied the effects of two D1 protein (Agrisera, Sweden). Horseradish peroxidase (HRP)- different types of herbicides, DCMU and bromoxynil. The results conjugated anti-chicken IgY was used as a secondary antibody. support the view that the acceptor-side mechanism is the major Chemiluminescent detection of HRP was carried out using ECL mechanism of photodamage to PSII under strong illumination. Western Blotting Detection Reagents (GE Healthcare, UK) and an imaging system with a CCD camera, Fluor-S multiImager (Bio-Rad, 2. Materials and methods Hercules, CA). The peak area corresponding to each band was quantified by ImageJ (National Institutes of Health, http://rsb.info. 2.1. Sample preparation and photodamage procedure nih.gov/ij/). It was verified that the assay provided a linear response over the range of 0.1–1.6 μg of Chl/lane of the control sample.

O2-evolving PSII membranes were prepared from spinach following the previous method [35] and suspended in a pH 6.5 buffer 3. Results containing 40 mM Mes, 400 mM sucrose and 20 mM NaCl (buffer A).

The PSII sample was suspended at the concentration of 0.17 mg Chl/ O2-evolving PSII membranes from spinach in the absence or mL in 3 mL of buffer A in the absence (as a control) or the presence of presence of herbicide (DCMU or bromoxynil) were subjected to 0.1 mM herbicide (DCMU or bromoxynil). The sample temperature strong illumination (2000 μEm− 2 s− 1) of white light (350–750 nm) was adjusted to 298 K by circulating water through a water jacket at 298 K for different periods (0, 30, 60, 90 and 120 min), and then the around a glass vessel (15 mm in diameter). The sample in this vessel activities of the three cofactors, QA, the Mn cluster, and P680, were was illuminated using a metal halide lamp (Luminar Ace LA-180 Me, examined after light-induced FTIR difference spectroscopy. Fig. 1 HAYASHI WATCH-WORKS, CO., LTD), which has a spectral range of shows FTIR difference spectra of the PSII samples after the photo- 350–750 nm and an intensity of 2000 μEm− 2 s− 1 at the sample damage treatment, which were measured by continuous illumination − position. The samples illuminated for 0 (no illumination), 30, 60, 90, at 210 K in the presence of DCMU. Under this condition, QA and 120 min were then washed with buffer A by centrifugation, and semiquinone anion is fully photo-accumulated due to the low stored in liquid nitrogen until FTIR measurements. temperature and DCMU to block the electron transfer beyond QA, − and hence FTIR signals representing the QA /QA difference should be 2.2. FTIR measurements detected. When QA is inactive, however, even if charge separation takes place, the resultant P680+Pheo− state quickly recombines and − Light-induced FTIR difference spectra were recorded on a Bruker no signal is detected. Thus, the intensity of the QA /QA signal directly IFS-66/S spectrophotometer equipped with an MCT detector (D313-L) represents the activity of QA, whereas the signals on the electron − 1 − at 4 cm resolution. For S2QA /S1QA difference measurement [36,37], donor side are detected only when QA is active. When the Mn cluster the PSII sample (0.5 mg Chl/mL) suspended in 1 mL of buffer A in the is active in QA-active centers, the S1 →S2 transition takes place and the presence of 0.1 mM DCMU was centrifuged at 170,000×g for 40 min S2/S1 difference signals are detected [36,37]. However, when the Mn and the resulting pellet was sandwiched between two circular BaF2 cluster is inactive, Cytb559, ChlZ or β-carotene in the secondary plates, each 13 mm in diameter. The sample temperature was electron pathway is oxidized instead of the Mn cluster and their adjusted to 210 K in a cryostat (Oxford DN1704) using a controller oxidized forms are accumulated at cryogenic temperatures [41–43]. 1216 I. Idedan et al. / Biochimica et Biophysica Acta 1807 (2011) 1214–1220

were different from that of the control sample, spectral features were virtually the same for all the samples. − − 1 In Fig. 2, the intensities of the QA /QA signal at 1479 cm (black − 1 symbols) and the S2/S1 signal at 1404 cm (blue symbols) in the above FTIR spectra (Fig. 1) are plotted as a function of illumination time during photodamage treatment. The error bars were estimated from the noise level at each position in the dark-minus-dark difference spectra. It is noticeable that the Mn-cluster signal (blue symbols) showed virtually the same illumination-time dependence

with the QA signal (black symbols) in each sample (circles: no herbicide; triangles: +DCMU; squares: +bromoxynil), reﬂecting the identical spectral features during the photodamage process shown in Fig. 1. This observation strongly suggests that the decrease in the Mn

cluster signal is ascribed to the inactivation of QA, and hence the rate of inactivation of the Mn cluster is similar to or slower, but at least not

faster, than that of QA. It was also clearly shown that the decreases in the QA and the Mn-cluster signals were slower and faster in the presence of DCMU (triangles) and bromoxynil (squares), respectively, than those in the absence of herbicide (circles). Fig. 3A shows the P680+/P680 FTIR difference spectra of control PSII membranes without herbicide subjected to strong illumination for 0, 30, 60, 90, and 120 min (from upper to lower spectra) at 298 K.

Before the measurements, the Mn cluster and QA were depleted by dithionite treatment [39], and hence the P680+/P680 signals represent the activities of only the cofactors involved in charge

separation, i.e., P680, ChlD1, and Pheo. A prominent positive doublet at 1723 and 1711 cm− 1 and a negative peak at 1702 cm− 1 in the P680+/P680 spectrum have been assigned to the keto C=O vibrations of P680+ and P680, respectively, and medium and weak peaks below 1600 cm− 1 to the chlorine ring vibrations [38]. The intensities of the spectra decreased with longer periods of illumina- − + Fig. 1. FTIR difference spectra representing QA /QA and S2/S1 differences of PSII tion without changing overall spectral features. Very similar P680 / membranes after photodamage treatment. The PSII samples in the absence of herbicide P680 spectra were observed for the PSII samples illuminated in the (A) and in the presence of DCMU (B) and bromoxynil (C) were subjected to strong presence of DCMU and bromoxynil (Fig. 3B and C, respectively). Note – μ − 2 − 1 illumination (350 750 nm; 2000 Em s ) for 0, 30, 60, 90, and 120 min (from top that some changes seen around 1650 cm-1 in the spectra at 90 min to bottom) at 298 K. Light-induced FTIR difference spectra were measured at 210 K in the presence of DCMU using continuous illumination. The spectra in each panel were with DCMU and at 120 min with bromoxynil are artifacts due to normalized based on the protein amount represented by the amide II band background disturbance in the amide I region where strong bands (~1550 cm− 1) intensities of the original (no difference) FTIR spectra. exist in the original (before taking difference) spectra. The intensity of the P680+/P680 FTIR signal, which was estimated as an intensity difference between the positive peak at 1723 cm− 1 Thus, the FTIR signals of either of these cofactors [36,44–46] together − with QA should be observed in the spectra. The upper spectrum in Fig. 1A, which was obtained using the control sample (in the absence of herbicide) without photodamage − treatment, showed typical S2QA /S1QA difference signals [36].The most prominent positive peak at 1479 cm− 1 arises from the CO/CC − stretching mode of QA semiquinone anion [47,48], while the negative peak at 1404 cm− 1 typically represents the symmetric − COO vibration of a carboxylate group coupled to the S1 → S2 transition of the Mn cluster [30,32,37]. Complex band features in the 1700–1600 cm− 1 are due mainly to the amide I bands (C_O stretches of backbone amides), while those in the 1600– 1500 cm− 1 arise from amide II bands (NH bend+CN stretches of backbone amides) and asymmetric COO− vibrations of carboxylate groups [30,32].Asthesamplewassubjectedtostrongilluminationat 298 K in a longer period, the overall spectral intensity decreased (Fig. 1A, upper to lower spectra). It is noteworthy, however, that the spectral features were basically identical even after long illumina- − tion, indicating that the S2QA /S1QA difference was always detected throughout the photodamage process. The absence of the typical − 1 + signals of Cytb559ox/Cytb559red at 1661/1655 cm [44],ChlZ /ChlZ Fig. 2. Illumination-time dependence of the FTIR signal intensities of QA (black − at 1713/1684 cm 1 [46],andβ-carotene cation at 1465 and symbols) and the Mn cluster (blue symbols) for PSII samples in the absence of herbicide 1441 cm− 1 [45] in the spectra of photodamaged samples (Fig. 1A) (circles) and in the presence of DCMU (triangles) and bromoxynil (squares). The FTIR intensities of QA and the Mn cluster were estimated as intensities of the positive peak at also supports the latter view. The corresponding spectra of the PSII − − 1479 cm 1 (ΔA difference from the 1494 cm 1 dip) and the negative peak at samples in the presence of DCMU and bromoxynil are presented in 1404 cm− 1 (ΔA difference from the 1416 cm− 1 dip), respectively, in the spectra in Fig. 1B and C, respectively. Although the rates of intensity decreases Fig. 1. I. Idedan et al. / Biochimica et Biophysica Acta 1807 (2011) 1214–1220 1217

but became larger at longer illumination times. This observation is in

contrast to that of the QA signal that showed nearly exponential decays (Fig. 2). The decrease in the P680 signal was slower and faster in the presence of DCMU (Fig. 4, triangles) and bromoxynil (Fig. 4, squares), respectively, than in the absence of herbicide (Fig. 4, circles).

This trend of the herbicide effect was similar to that of the QA and Mn cluster signals (Fig. 2). The photodamage process was further analyzed by estimating the degree of degradation of the D1 polypeptide by strong illumination. Fig. 5 shows the image of Western blotting using an anti-D1 C- terminal domain antibody for PSII samples in the absence of herbicide (top panel) and in the presence of DCMU (middle panel) and bromoxynil (bottom panel) before and after illumination for 60 min. The amount of the D1 protein in the samples without herbicide and with DCMU and bromoxynil decreased to 80, 84 and 47%, respectively, after 60 min illumination. Thus, degradation of the D1 protein was also slower and faster in the presence of DCMU and bromoxynil, respectively, in comparison with that in the absence of herbicide.

In Fig. 6, the time course of the decreases in the QA (closed circles) and P680 (open circles) signals and of the D1 degradation (crosses) is plotted together for each sample (A: no herbicide; B: +DCMU; C: + bromoxynil). It is shown that the decrease in the P680 signal was

always slower than the decay of the QA signal irrespective of the herbicide condition. Also, degradation of the D1 protein was even slower than the decrease in the P680 signal in all the samples.

4. Discussion

In this study, herbicide effects on the photodamage to PSII membranes from spinach by strong white-light illumination (350– 750 nm; 2000 μEm− 2 s− 1) were examined using light-induced FTIR + Fig. 3. P680 /P680 FTIR difference spectra of PSII membranes after photodamage difference spectroscopy to obtain insight into the molecular mecha- treatment. The PSII samples in the absence of herbicide (A) and in the presence of DCMU (B) and bromoxynil (C) were subjected to strong illumination (350–750 nm; nism of photodamage. With this technique, we can directly monitor 2000 μEm− 2 s− 1) for 0, 30, 60, 90, and 120 min (from top to bottom) at 298 K. The the activities of individual redox cofactors. In the control PSII photodamaged samples were depleted of QA by dithionite treatment and light-induced membranes in the absence of herbicide and in the presence of P680+/P680 difference spectra were measured at 265 K in the presence of ferricyanide DCMU and bromoxynil, the QA activities as measured by the FTIR and SiMo. The spectra in each panel were normalized based on the protein amount − − 1 − signal of the Q semiquinone anion at 1479 cm (Fig. 1) decreased represented by the amide II band (~1550 cm 1) intensities of the original (no A difference) FTIR spectra. in the course of illumination (Fig. 2, black symbols). The decay rates were signiﬁcantly inﬂuenced by the presence of herbicides; DCMU

andbromoxynilretardedandacceleratedtheQA inactivation, and the negative peak at 1702 cm− 1, was plotted as a function of respectively (Fig. 2, black triangles and squares, respectively). This illumination time in Fig. 4. The intensity decrease showed a kind of tendency of herbicide effect is in agreement with the previous reports sigmoidal curve; the decay rate was relatively small in the early stage on the effect of urea- and phenol-type herbicides on the inactivation of electron transfer reactions [25–28]. − 1 The decrease in the S2/S1 FTIR signal at 1404 cm (Fig. 1)by strong illumination exhibited decay curves virtually identical to the

QA signal decays irrespective of herbicide conditions (Fig. 2, blue 60 min 0 min

Control

+DCMU

+Bromoxynil

Fig. 4. Illumination-time dependence of the FTIR signal intensities of P680 for PSII samples in the absence of herbicide (circles) and in the presence of DCMU (triangles) Fig. 5. Degradation of the D1 polypeptide by strong illumination detected by Western and bromoxynil (squares). The FTIR intensity of P680 was estimated as a ΔA difference blotting using an anti-D1 C-terminal domain antibody. PSII membranes without between the positive and negative peaks at 1723 and 1702 cm− 1, respectively, in the herbicide (control) and in the presence of DCMU and bromoxynil were subjected to P680+/P680 spectra in Fig. 3. strong illumination (350–750 nm; 2000 μEm− 2 s− 1) for 0 and 60 min at 298 K. 1218 I. Idedan et al. / Biochimica et Biophysica Acta 1807 (2011) 1214–1220

dently of electron transfer reactions. Hence, in this mechanism, inactivation of the Mn cluster should be dependent on neither herbicide binding nor its species. Thus, the observations of herbicide

dependence of the QA and the Mn cluster signals indicate that QA, but not the Mn cluster, is initially inactivated in the PSII photodamage process under the present experimental condition. The illumination-induced decrease in the P680 FTIR signal (Fig. 3)

also showed herbicide dependence similar to the QA signal, i.e., slower and faster decreases in the presence of DCMU and bromoxynil, respectively (Fig. 4). However, there was always some delay in the

decrease in the P680 signal in comparison with that in the QA signal (Fig. 7), and the decay curve of P680 showed a kind of sigmoidal shape indicative of the presence of a precedent reaction that triggers the P680 inactivation. The loss of the P680 FTIR signal represents the inactivation of redox components involved in primary charge

separation, i.e., P680, ChlD1 and Pheo [52]. Thus, the slower decrease in the P680 signal is in agreement with the previous observations that − 2+ the loss of Pheo reduction was slower than that of the QA Fe EPR signal and inhibition of the Hill reaction in the photodamage process [50,51,53,54]. These data in the present and previous studies suggest

that QA inactivation triggers the damage of P680, Pheo, or ChlD1. All of these data are consistent with the acceptor side mechanism

of PSII photodamage [9], i.e., double reduction of QA by strong illumination causes a relatively long lifetime (~1 ms) of the triplet 1 1 state on ChlD1 that induces the formation of harmful O2*. This O2* further damages other cofactors and the D1 protein. The observation of slower D1 degradation than the P680 inactivation (Figs. 5 and 6)is consistent with the view that the conformational change of the D1 protein induced by the damage to the charge separation cofactors

(P680/ChlD1/Pheo) or the nearby protein moiety triggers the D1 degradation [50].

The herbicide effect on the rate of double reduction of QA may be explained by the redox potential change of QA (Fig. 7). It is known that − DCMU and bromoxynil increase and decrease, respectively, the QA /QA redox potential [21]. This is due probably to the interaction transfer

from the QB site to QA through the QA–His–Fe–His–QB (QB =PQ or herbicide) bridge [55]. Indeed, hydrogen bond changes in the CO bond − of the QA semiquinone anion have been detected in the FTIR difference study [48], in which phenol-type herbicides and other types of herbicides (urea- , triazine, and uracil-type herbicides) − shifted the CO/CC stretching band of QA to a lower and higher frequency, respectively. The effects of the hydrogen bond change at

the conjugated C_O bond on the structure and the energy of QA are expected to be larger in its anionic state than the neutral state (Fig. 7), in analogy to the hydrogen bond effects on Pheo and Pheo−, which were recently demonstrated in the study using FTIR spectroscopy and density functional theory calculation [56]. Thus, from the above

Bx G(QA QA ) Fig. 6. Comparison of the relative amplitudes of active QA (closed circles) and P680 Q (open circles) and the D1 protein amount (crosses) as a function of illumination time B Bx G(QA PQH2) for PSII membranes in the absence of herbicides (A) and in the presence of DCMU (B) DCMU and bromoxynil (C). QB Q Free Energy Free A symbols). As for the PSII membranes in the absence of herbicide, it has DCMU already been shown that the decreases in the O2 evolution and the QA activity by white-light illumination provided similar decay curves Q [49–51]. Kirilovsky et al. [25] also showed that the extent of the losses A PQH2 − 2+ of O2 evolution and the QA Fe EPR signal after 30 min illumination was similar to each other both in the absence and in the presence of

DCMU with smaller losses in the latter case. The virtually identical Em(QA /QA) Em(PQH2/QA ) decays of the QA and Mn cluster signals indicate that the Mn cluster is DCMU> QB >Bx Bx > QB > DCMU inactivated not faster than QA, because the presence of active QA is prerequisite for the capability of the S1 →S2 transition or O2 evolution. − Fig. 7. Schematic diagram of relationships between the energy levels of QA,QA , and In the Mn mechanism of photodamage [6,7,18,19], the Mn cluster is − − PQH2 and the redox potentials of the QA /QA and PQH2/QA couples in the absence of thought to be inactivated directly by its light absorption indepen- herbicide (QB), and in the presence of DCMU and bromoxynil (Bx). See text for details. I. Idedan et al. / Biochimica et Biophysica Acta 1807 (2011) 1214–1220 1219

− herbicide effect on the QA /QA redox potential [21], it is predicted that 21570038 and 22370017 to T.T.) and by the grant from JST PRESTO − the energy level of QA decreases and increases by binding of DCMU program (T.T.). and bromoxynil, respectively (Fig. 7). Since PQH2 out of the QA pocket is independent of herbicide conditions, the redox potential of the − PQH2/QA couple should decrease and increase by DCMU and References bromoxynil, respectively. Thus, PQH2 formation by overreduction of — − [1] J. Barber, B. Andersson, Too much of a good thing light can be bad for QA will be suppressed by DCMU and accelerated by bromoxynil. photosynthesis, Trends Biochem. Sci. 17 (1992) 61–66. So far, the general tendency of the herbicide effect on the [2] E.-M. Aro, I. Virgin, B. Andersson, Photoinhibition of photosystem II. Inactivation, – photodamage to PSII, i.e., slower and faster PSII degradation by protein damage and turnover, Biochim. Biophys. Acta 1143 (1993) 113 134. [3] N. Adir, H. Zer, S. Shochat, I. Ohad, Photoinhibition — a historical perspective, DCMU and bromoxynil, respectively [22–29], has been explained in Photosynth. Res. 76 (2003) 343–370. + − the context of charge recombination between P680 and QA [5,8]. [4] I. Vass, E.-M. Aro, Photoinhibition of photosynthetic electron transport, in: G. + − Renger (Ed.), Primary Processes of Photosynthesis: Principles and Apparatus. Direct charge recombination of P680 QA competes with charge 3 + − 3 Comprehensive Series in Photochemical and Photobiological Sciences, The Royal recombination through [P680 Pheo ] to form ChlD1*, which is the Society of Chemistry, Cambridge, 2008, pp. 393–425. 1 precursor of harmful O2*. According to the previous model [5,8], the [5] A. Krieger-Liszkay, C. Fufezan, A. Trebst, Singlet oxygen production in photosys- − tem II and related protection mechanism, Photosynth. Res. 98 (2008) 551–564. downshift of the QA /QA redox potential by bromoxynil decreases the [6] S. Takahashi, N. Murata, How do environmental stresses accelerate photoinhibi- energy gap between Pheo and QA to facilitate charge recombination tion? Trends Plant Sci. 13 (2008) 178–182. through a triplet state accelerating photodamage, whereas the upshift [7] E. Tyystjärvi, Photoinhibition of photosystem II and photodamage of the oxygen − evolving manganese cluster, Coord. Chem. Rev. 252 (2008) 361–376. of the QA /QA redox potential by DCMU enhances the rate of the direct [8] I. Vass, K. Cser, Janus-faced charge recombinations in photosystem II photoinhibi- charge recombination pathway suppressing photodamage. Because tion, Trends Plant Sci. 14 (2009) 200–205. 1 3 the photodamage is initiated by O2* formed from ChlD1* in this [9] I. Vass, S. Styring, T. Hundal, A. Koivuniemi, E.-M. Aro, B. Andersson, Reversible and irreversible intermediates during photoinhibition of photosystem II: stable mechanism, the most possible site of initial damage is ChlD1 itself or the adjacent cofactors, P680 and Pheo. Hence, it is expected that reduced QA species promote chlorophyll triplet formation, Proc. Nat. Acad. Sci. U. S. A. 89 (1992) 1408–1412. + inactivation of P680 formation by the damage to ChlD1, P680 or Pheo [10] F. van Mieghem, K. Brettel, B. Hillmann, A. Kamlowski, A.W. Rutherford, E. takes place faster than or at least in the same timing as the Schlodder, Charge recombination reactions in photosystem II. 1. Yields, inactivation of Q . The observed delay of the P680 inactivation recombination pathways, and kinetics of the primary pair, Biochemistry 34 A (1995) 4798–4813. compared with that of the QA inactivation (Fig. 6) is however [11] T. Noguchi, T. Tomo, C. Kato, Triplet formation on a monomeric chlorophyll in the contradictory to this prediction. Thus, although the above mechanism photosystem II reaction center as studied by time-resolved infrared spectroscopy, – is suitable under low or medium light conditions, it is highly likely Biochemistry 40 (2001) 2176 2185. μ − 2 − 1 [12] N. Keren, A. Berg, P.J.M. van Kan, H. Levanon, I. Ohad, Mechanism of photosystem II that under the strong light condition (~2000 Em s ) used in the photoinactivation and D1 protein degradation at low light: the role of back present study, the observed herbicide effect is caused by the different electron flow, Proc. Nat. Acad. Sci. U. S. A. 94 (1997) 1579–1584. [13] N. Keren, H. Gong, I. Ohad, Oscillations of reaction center II-D1 protein degradation in rates of double reduction of QA. - vivo induced by repetitive light flashes. Correlation between the level of RCII-QB and The delay of inactivation of P680 in comparison with QA in all the protein-degradation in low light, J. Biol. Chem. 270 (1995) 806–814. samples without and with herbicides (Fig. 6) is also consistent with [14] A. Szilárd, L. Sass, E. Hideg, I. Vass, Photoinactivation of Photosystem II by flashing light, Photosynth. Res. 84 (2005) 15–20. the triplet quenching phenomena when QA is singly reduced [10], − [15] D.J. Blubaugh, G.M. Cheniae, Kinetics of photoinhibition in hydroxylamine- because QA should be accumulated during strong illumination before extracted photosystem II membranes: relevance to photoactivation and sites of 3 its double reduction. It has been observed that ChlD1* decays ~100 electron donation, Biochemistry 29 (1990) 5109–5118. − [16] G.X. Chen, J. Kazimir, G.M. Cheniae, Photoinhibition of hydroxylamine-extracted times faster in the presence of QA (t1/2 ~10 μs) than in the case with photosystem-II membranes: studies of the mechanism, Biochemistry 31 (1992) an empty QA pocket by its double reduction (t1/2 ~1 ms) [10]. This – − 11072 11083. triplet quenching by QA has been explained by electron transfer [17] C. Jegerschöld, I. Virgin, S. Styring, Light-dependent degradation of the D1 protein reactions via 3Pheo* [10] or triplet transfer to a β-carotene molecule in photosystem II is accelerated after inhibition of the water splitting reaction, – via the D2 branch [57]. Although the exact molecular mechanism of Biochemistry 29 (1990) 6179 6186. [18] M. Hakala, I. Tuominen, M. Keränen, T. Tyystjärvi, E. Tyystjärvi, Evidence for the triplet quenching is still unknown, the present results of the P680 role of the oxygen-evolving manganese complex in photoinhibition of photosys- − – inactivation delays (Fig. 6) provide experimental evidence that QA tem II, Biochim. Biophys. Acta 1706 (2005) 68 80. has a photoprotection function against PSII damage [58]. When Q− is [19] N. Ohnishi, S.I. Allakhverdiev, S. Takahashi, S. Higashi, M. Watanabe, Y. Nishiyama, A N. Murata, Two-step mechanism of photodamage to photosystem II: step 1 occurs accumulated under strong illumination, PSII can escape from photo- at the oxygen-evolving complex and step 2 occurs at the photochemical reaction – damage by rapid triplet quenching. If QA is doubly reduced by further center, Biochemistry 44 (2005) 8494 8499. 1 [20] E. Tyystjärvi, E.-M. Aro, The rate constant of photoinhibition, measured in illumination, then P680 and the D1 protein are degraded by O2* 3 lincomycin-treated leaves, is directly proportional to light intensity, Proc. Nat. formed from long-lived ChlD1*. Acad. Sci. U. S. A. 93 (1996) 2213–2218. In conclusion, the results of herbicide effects in the present [21] A. Krieger-Liszkay, A.W. Rutherford, Influence of herbicide binding on the redox study strongly support the view that the acceptor-side mechanism potential of the quinone acceptor in photosystem II. Relevance to photodamage and phytotoxicity, Biochemistry 37 (1998) 17339–17344. dominantly occurs in the photodamage to PSII by strong white-light [22] A. Trebst, B. Depka, B. Kraft, U. Johanningmeier, The QB site modulates the illumination. The light used in this study (350–750 nm) includes a conformation of the photosystem II reaction center polypeptides, Photosynth. Res. part of UV-A (315–400 nm) in addition to the visible region. 18 (1988) 163–177. Because it has been observed that UV and blue light induce [23] M. Kuhn, P. Böger, Studies on the light-induced loss of the D1 protein in photosystem-II membrane fragments, Photosynth. Res. 23 (1990) 291–296. inactivation of the Mn cluster [18,19,59,60],photodamagetoPSII [24] M.A.K. Jansen, B. Depka, A. Trebst, M. Edelman, Engagement of specific sites in the might occur as mixed events by different mechanisms [61,62]. plastoquinone niche regulates degradation of the D1 protein in photosystem II, J. – However, the data in the present study showed that under strong Biol. Chem. 268 (1993) 21246 21252. [25] D. Kirilovsky, A.W. Rutherford, A.L. Etienne, Influence of DCMU and ferricyanide white light, QA is still the main target of initial photodamage among on photodamage in photosystem II, Biochemistry 33 (1994) 3087–3095. PSII cofactors. [26] H. Zer, I. Ohad, Photoinactivation of photosystem II induces changes in the photochemical reaction center II abolishing the regulatory role of the QB site in the D1 protein degradation, Eur. J. Biochem. 231 (1995) 448–453. [27] Y. Nakajima, S. Yoshida, T. Ono, Differential effects of urea triazine-type and phenol-type Acknowledgements photosystem II inhibitors on inactivation of the electron transport and degradation of the D1 protein during photoinhibition, Plant Cell Physiol. 37 (1996) 673–680. This study was supported by the Grants-in-Aid for Scientific [28] J. Komenda, M. Koblížek, O. Prášil, Characterization of processes responsible for the distinct effect of herbicides DCMU and BNT on Photosystem II photoinactiva- Research from the Ministry of Education, Culture, Sports, Science and tion in cells of the cyanobacterium Synechococcus sp PCC 7942, Photosynth. Res. Technology (21370063, 23108706, and 23657099 to T.N. and 63 (2000) 135–144. 1220 I. Idedan et al. / Biochimica et Biophysica Acta 1807 (2011) 1214–1220

[29] C. Fufezan, A.W. Rutherford, A. Krieger-Liszkay, Singlet oxygen production in [47] C. Berthomieu, E. Nabedryk, W. Mäntele, J. Breton, Characterization by FTIR

herbicide-treated photosystem II, FEBS Lett. 532 (2002) 407–410. spectroscopy of the photoreduction of the primary quinone acceptor QA in [30] T. Noguchi, C. Berthomieu, Molecular analysis by vibrational spectroscopy, in: T. photosystem II, FEBS Lett. 269 (1990) 363–367. Wydrzynski, K. Satoh (Eds.), Photosystem II: The Light-Driven Water: Plastoqui- [48] A. Takano, R. Takahashi, H. Suzuki, T. Noguchi, Herbicide effect on the hydrogen-

none Oxidoreductase, Springer, Dordrecht, The Netherland, 2005, pp. 367–387. bonding interaction of the primary quinone electron acceptor QA in photosystem [31] T. Noguchi, Light-induced FTIR difference spectroscopy as a powerful tool toward II as studied by Fourier transform infrared spectroscopy, Photosynth. Res. 98 understanding the molecular mechanism of photosynthetic oxygen evolution, (2008) 159–167. Photosynth. Res. 91 (2007) 59–69. [49] T. Ono, T. Noguchi, Y. Nakajima, Characteristic changes of function and structure of [32] T. Noguchi, Fourier transform infrared analysis of the photosynthetic oxygen- photosystem II during strong-light photoinhibition under aerobic conditions, evolving center, Coord. Chem. Rev. 252 (2008) 336–346. Biochim. Biophys. Acta 1229 (1995) 239–248. [33] R.J. Debus, Protein ligation of the photosynthetic oxygen-evolving center, Coord. [50] K.J. van Wijk, B. Andersson, S. Styring, Spectroscopic characterization of Chem. Rev. 252 (2008) 244–258. photoinhibited photosystem II and kinetic resolution of the triggering of the D1 [34] C. Berthomieu, R. Hienerwadel, Fourier transform infrared (FTIR) spectroscopy, reaction center protein for degradation, Biochim. Biophys. Acta 1100 (1992) Photosynth. Res. 101 (2009) 157–170. 207–215. [35] T. Ono, Y. Inoue, Effects of removal and reconstitution of the extrinsic 33, 24 and [51] S. Styring, I. Virgin, A. Ehrenberg, B. Andersson, Strong light photoinhibition of 16 kDa proteins on ﬂash oxygen yield in photosystem II particles, Biochim. electron-transport in photosystem II. Impairment of the function of the ﬁrst

Biophys. Acta 850 (1986) 380–389. quinone acceptor, QA, Biochim. Biophys. Acta 1015 (1990) 269–278. [36] T. Noguchi, T. Ono, Y. Inoue, Temperature-dependence of the S1 →S2 transition in [52] B.A. Diner, F. Rappaport, Structure, dynamics, and energetics of the primary the oxygen-evolving complex of photosystem II studied by FT-IR spectroscopy, photochemistry of photosystem II of oxygenic photosynthesis, Annu. Rev. Plant Biochim. Biophys. Acta 1143 (1993) 333–336. Biol. 53 (2002) 551–580. [37] T. Noguchi, T. Ono, Y. Inoue, Direct detection of a carboxylate bridge between Mn [53] S.I. Allakhverdiev, E. Šetliková, V.V. Klimov, I. Šetlík, In photoinhibited photosys- and Ca2+ in the photosynthetic oxygen-evolving center by means of Fourier tem II particles pheophytin photoreduction remains unimpaired, FEBS Lett. 226 transform infrared spectroscopy, Biochim. Biophys. Acta 1228 (1995) 189–200. (1987) 186–190. [38] T. Okubo, T. Tomo, M. Sugiura, T. Noguchi, Perturbation of the structure of P680 [54] L. Nedbal, J. Masojídek, J. Komenda, O. Prášil, I. Šetlík, Three types of photosystem and the charge distribution on its radical cation in isolated reaction center II photoinactivation. 2. Slow processes, Photosynth. Res. 24 (1990) 89–97. complexes of photosystem II as revealed by Fourier transform infrared [55] R. Takahashi, K. Hasegawa, A. Takano, T. Noguchi, The structures and binding sites

spectroscopy, Biochemistry 46 (2007) 4390–4397. of phenolic herbicides in the QB pocket of photosystem II, Biochemistry 49 (2010) [39] F.J.E. van Mieghem, W. Nitschke, P. Mathis, A.W. Rutherford, The inﬂuence of the 5445–5454. quinone-iron electron acceptor complex on the reaction centre photochemistry of [56] Y. Shibuya, R. Takahashi, T. Okubo, H. Suzuki, M. Sugiura, T. Noguchi, Hydrogen photosystem II, Biochim. Biophys. Acta 977 (1989) 207–214. bond interactions of the pheophytin electron acceptor and its radical anion in [40] M. Ikeuchi, Y. Inoue, A new 4.8 kDa polypeptide intrinsic to the PSII reaction photosystem II as revealed by Fourier transform infrared difference spectroscopy, center, as revealed by modiﬁed SDS-PAGE with improved resolution of low- Biochemistry 49 (2010) 493–501. molecular weight proteins, Plant Cell Physiol. 29 (1988) 1233–1239. [57] V. Martínez-Junza, M. Szczepaniak, S.E. Braslavsky, J. Sander, M. Nowaczyk, M.

[41] D.H. Stewart, G.W. Brudvig, Cytochrome b559 of photosystem II, Biochim. Biophys. Rögner, A.R. Holzwarth, A photoprotection mechanism involving the D2 branch in Acta 1367 (1998) 63–87. photosystem II cores with closed reaction centers, Photochem. Photobiol. Sci. 7 [42] C.A. Tracewell, J.S. Vrettos, J.A. Bautista, H.A. Frank, G.W. Brudvig, Carotenoid (2008) 1337–1343. photooxidation in photosystem II, Arch. Biochem. Biophys. 385 (2001) 61–69. [58] T. Noguchi, Dual role of triplet localization on the accessory chlorophyll in

[43] P. Faller, C. Fufezan, A.W. Rutherford, Side-path electron donors: Cytochrome b559, photosystem II reaction center: photoprotection and photodamage to the D1 chlorophyll Z and β-carotene, in: T. Wydrzynski, K. Satoh (Eds.), Photosystem II: protein, Plant Cell Physiol. 43 (2002) 1112–1116. The Light-Driven Water:Plastoquinone Oxidoreductase, Springer, Dordrecht, The [59] G. Renger, M. Völker, H.J. Eckert, R. Fromme, S. Hohmveit, P. Gräber, On the Netherlands, 2005, pp. 347–365. mechanism of photosystem II deterioration by UV-B irradiation, Photochem. [44] C. Berthomieu, A. Boussac, W. Mäntele, J. Breton, E. Nabedryk, Molecular changes Photobiol. 49 (1989) 97–105. following oxidoreduction of cytochrome b559 characterized by Fourier transform [60]I.Vass,E.Turcsanyi,E.Touloupakis,D.Ghanotakis,V.Petrouleas,The infrared spectroscopy and electron paramagnetic resonance: photooxidation in mechanism of UV-A radiation-induced inhibition of photosystem II electron photosystem II and electrochemistry of isolated cytochrome b559 and iron transport studied by EPR and chlorophyll ﬂuorescence, Biochemistry 41 protoporphyrin IX-bisimidazole model compounds, Biochemistry 31 (1992) (2002) 10200–10208. 11460–11471. [61] H.-J. Eckert, B. Geiken, J. Bernarding, A. Napiwotzki, H.-J. Eichler, G. Renger, Two sites [45] T. Noguchi, T. Mitsuka, Y. Inoue, Fourier transform infrared spectrum of the radical of photoinhibition of the electron transfer in oxygen evolving and Tris-treated PS II cation of β-carotene photoinduced in photosystem II, FEBS Lett. 356 (1994) 179–182. membrane fragments from spinach, Photosynth. Res. 27 (1991) 97–108. [46] Y. Kitajima, T. Noguchi, Photooxidation pathway of chlorophyll Z in photosystem [62] R. Oguchi, I. Terashima, W.S. Chow, The involvement of dual mechanisms of II as studied by Fourier transform infrared spectroscopy, Biochemistry 45 (2006) photoinactivation of photosystem II in Capsicum annuum L. plants, Plant Cell 1938–1945. Physiol. 50 (2009) 1815–1825.