Physiologia Plantarum 143: 396–407. 2011 Copyright © Physiologia Plantarum 2011, ISSN 0031-9317

Mitochondrial alternative oxidase pathway protects plantsagainst photoinhibition by alleviating inhibition of therepair of photodamaged PSII through preventing formationof reactive oxygen species in Rumex K-1 leavesLi-Tao Zhang, Zi-Shan Zhang, Hui-Yuan Gao∗, Zhong-Cai Xue, Cheng Yang, Xiang-Long Mengand Qing-Wei Meng

State Key Laboratory of Crop Biology, Shandong Key Laboratory of Crop Biology, Shandong Agricultural University, Tai’an 271018,Shandong, China

Correspondence*Corresponding author,e-mail: gaohy@sdau.edu.cn

Received 18 May 2011;revised 21 July 2011

doi:10.1111/j.1399-3054.2011.01514.x

The purpose of this study was to explore how the mitochondrial AOX(alternative oxidase) pathway alleviates photoinhibition in Rumex K-1 leaves.Inhibition of the AOX pathway decreased the initial activity of NADP-malate dehydrogenase (EC 1.1.1.82, NADP-MDH) and the pool size ofphotosynthetic end electron acceptors, resulting in an over-reduction of thephotosystem I (PSI) acceptor side. The over-reduction of the PSI acceptor sidefurther inhibited electron transport from the photosystem II (PSII) reactioncenters to the PSII acceptor side as indicated by an increase in VJ (the relativevariable fluorescence at J-step), causing an imbalance between photosyntheticlight absorption and energy utilization per active reaction center (RC) underhigh light, which led to the over-excitation of the PSII reaction centers.The over-reduction of the PSI acceptor side and the over-excitation of thePSII reaction centers enhanced the accumulation of reactive oxygen species(ROS), which inhibited the repair of the photodamaged PSII. However, theinhibition of the AOX pathway did not change the level of photoinhibitionunder high light in the presence of the chloroplast D1 protein synthesisinhibitor chloramphenicol, indicating that the inhibition of the AOX pathwaydid not accelerate the photodamage to PSII directly. All these results suggestthat the AOX pathway plays an important role in the protection of plantsagainst photoinhibition by minimizing the inhibition of the repair of thephotodamaged PSII through preventing the over-production of ROS.

Abbreviations – φEo, the quantum yield of electron transport; φPo, the maximum quantum yield for primary photochemistry;�PSII, photosystem II actual photochemical efficiency; �0, the efficiency for electron transport; ABS/RC, light absorption peractive reaction center; AOX, alternative oxidase; APX, ascorbate peroxidase; CAT, catalase; CM, chloramphenicol; COX,cytochrome oxidase; DAB,3, 3-diaminobenzidine; DTT, dithiothreitol; ET0/RC, the electron transport per active reactioncenter; FJ, the fluorescence at J-step; Fo, Fm, initial and maximum fluorescence; F′

o, F′m, the minimal and maximum

fluorescence in the light-adapted state; FS, the steady-state fluorescence; Ft, the fluorescence at any time t; HSD, honestlysignificant difference; LED, light-emitting diode; LSD, least significant difference; NADP/NADPH, oxidized/reduced formof nicotinamide-adenine dinucleotide phosphate; NADP-MDH, NADP-malate dehydrogenase; NBT, nitrotetrazolium bluechloride; nPG, n-propyl gallate; OAA, oxaloacetate; PFD, photon flux density; PSI, photosystem I; PSII, photosystem II; QA,primary quinone electron acceptor of PS II; qP, photochemical quenching; RC, active reaction center; ROS, reactive oxygenspecies; SHAM, salicylhydroxamic acid; SOD, superoxide dismutase; TR0/RC, trapping of excitation energy per active reactioncenter; VJ, the relative variable fluorescence at the J-step.

396 Physiol. Plant. 143, 2011

Introduction

Excess light energy damages the photosynthetic machin-ery during photosynthesis, causing photoinhibition(Murata et al. 2007, Takahashi and Murata 2008). It hasbeen demonstrated that photodamage is attributable tolight absorbed directly by manganese in the oxygen-evolving complex (Nishiyama et al. 2006, Tyystjarvi2008). Subsequent to photodamage to the oxygen-evolving complex, the reaction center of photosystemII (PSII) is damaged by light absorbed by the photosyn-thetic pigments (Hakala et al. 2005, Ohnishi et al. 2005).The damaged PSII is rapidly and effectively repairedby the replacement of the damaged PSII proteins withnewly synthesized proteins, primarily the D1 protein(Aro et al. 2005, Mattoo and Edelman 1987, Nishiyamaet al. 2011). Thus, photoinhibition occurs only underconditions in which the rate of photodamage exceedsthe rate of its repair (Murata et al. 2007, Nishiyama et al.2011, Takahashi and Murata 2008). To prevent photoin-hibition, plants have evolved mechanisms to suppressphotodamage to PSII and/or to minimize the inhibition ofthe repair of the photodamaged PSII. It has been demon-strated that the photorespiratory pathway and antioxi-dant systems protect plants by minimizing the inhibitionof the repair of the photodamaged PSII (Nishiyama et al.2001, Takahashi et al. 2007), and chloroplast avoid-ance movement can help to prevent photoinhibitionby directly suppressing the photodamage to PSII (Kasa-hara et al. 2002). Furthermore, Takahashi et al. (2009)suggested that photosynthetic cyclic electron flow canprotect plants against photoinhibition by both prevent-ing photodamage to PSII and minimizing the inhibitionof the repair of the photodamaged PSII.

Recent publications have also suggested that the mito-chondrial alternative oxidase (AOX) pathway plays animportant role in protecting photosynthetic machineryfrom photoinhibition (Florez-Sarasa et al. 2011, Noguchiand Yoshida 2008, Raghavendra and Padmasree 2003).The AOX pathway is an alternative electron transportpathway in addition to the cytochrome oxidase (COX)pathway in the mitochondria. This pathway diverts elec-trons from the ubiquinone pool and reduces oxygento water without any proton translocation or ATP syn-thesis (Rasmusson et al. 2009, Van Aken et al. 2009).Apparently, the AOX pathway is non-phosphorylatingand can oxidize reducing equivalents efficiently withoutbeing restricted by the proton gradient across the mito-chondrial inner membrane or the cellular ATP/ADP ratio(Yoshida et al. 2007). Several recent studies have demon-strated that the inhibition of the AOX pathway leads todecreases in photosynthetic rate in leaves (Yoshida et al.2006) and protoplasts (Dinakar et al. 2010a, 2010b,

Padmasree and Raghavendra 1999a, 1999c). It has beenshown that the AOX pathway is induced to protect thephotosynthetic apparatus from photoinhibition in Ara-bidopsis, comprising the cyclic electron flow aroundphotosystem I (PSI) (Yoshida et al. 2007). However, itremains to be elucidated whether the photoprotection ofthe mitochondrial AOX pathway is correlated with thesuppression of photodamage to PSII and/or minimizingthe inhibition of the repair of the photodamaged PSII.

The accumulation of ROS induced by excess lightenergy has been suggested to play an important roleduring photoinhibition (Murata et al. 2007, Takahashiand Murata 2008). A recent study, using plant materiallacking AOX1a, demonstrated that the AOX pathwayplays a central role in reducing the accumulation ofROS generated by photosynthesis in plants (Giraudet al. 2008). However, the generation site of ROS inchloroplasts and the role of ROS accumulation duringphotoinhibition conditions because of the inhibition ofAOX pathway have not yet been clarified.

The aim of this study was to explore the role of ROSaccumulation resulting from the inhibition of the AOXpathway during photoinhibition and to clarify whetherthe photoprotection of the mitochondrial AOX pathwayis correlated with the suppression photodamage to PSIIand/or with minimizing the inhibition of the repair of thephotodamaged PSII. In addition, using a chlorophyll afluorescence transient combined with the JIP-test (Jiaet al. 2010, Kalachanis and Manetas 2010, Mathuret al. 2011, Oukarroum et al. 2009, Strasser et al. 2000,2004), we studied photosynthetic behaviors (includingthe energy fluxes of absorption, energy trapping andelectron transport) to analyze the generation site of ROSin chloroplasts under high light when the AOX pathwayis inhibited.

Materials and methods

Materials and treatments

Rumex K-1 plants (Rumex patientia × Rumex tian-schaious) were grown from seeds under a 14-h photope-riod at 22/18◦C (day/night) in a pot containing soil. Theplants were thinned to one plant per pot 2 weeks aftersowing. Nutrients and water were supplied sufficientlythroughout the study to avoid any potential nutrient anddrought stresses. The photon flux density (PFD) dur-ing growth was approximately 800 μmol m−2 s−1. Fullyexpanded leaves of 4-week-old plants were used in theexperiments.

Intact chloroplasts were isolated according to theprotocol of Bartoli et al. (2005). Leaves were groundin buffer containing 50 mM HEPES-KOH (pH 7.5),

Physiol. Plant. 143, 2011 397

330 mM sorbitol, 2 mM Na2-EDTA, 1 mM MgCl2,5 mM ascorbic acid and 0.05% (w/v) BSA using ahand-held homogenizer. The homogenate was filteredthrough a 20 μm pore size nylon mesh and centrifugedat 3000 g for 5 min. The pellet was suspended inwash medium (50 mM HEPES-KOH pH 7.5 and330 mM sorbitol), loaded in a solution consisting of35% (v/v) Percoll in wash medium and centrifugedat 2500 g for 5 min. The pellet containing intactchloroplasts was used to measure modulated chlorophyllfluorescence in the absence (control) or presenceof 1 mM salicylhydroxamic acid (SHAM) or n-propylgallate (nPG). The isolated chloroplasts were 75% intactaccording to the ferricyanide-dependent O2 evolution(Walker, 1988), which was measured with a Chlorolab-2oxygen electrode (Hansatech Instruments, Norfolk, UK).

Leaf discs (0.5 cm2) were punched from fullyexpanded leaves and infiltrated with 0 (control) or 1 mMSHAM or nPG solutions, respectively, for 2 h in the darkat room temperature. Then, the discs were exposed tohigh light measuring 800 μmol m−2 s−1 for 1 h at roomtemperature.

Capacity of AOX pathway measurement

The respiratory rate in the leaf discs treated with highlight was measured using an Oxytherm oxygen electrode(Hansatech Instruments, Norfolk, UK) at 25◦C accordingto Yoshida et al. (2006) and Zhang et al. (2010). Theleaf discs were incubated in the dark for 10 min beforerespiratory measurements were taken. The inhibitors1 mM KCN and 20 mM SHAM were used to inhibitthe COX and AOX pathways, respectively. The AOXpathway capacity was defined as the O2 uptake rate inthe presence of KCN that was sensitive to SHAM.

Enzyme assays

Enzymes were extracted from the leaves according toDutilleul et al. (2003). Leaf discs (7.5 cm2) were groundin liquid nitrogen and extracted in 50 mM HEPES-KOH(pH 7.5) buffer containing 10 mM MgCl2, 1 mM Na2-EDTA, 5 mM dithiothreitol (DTT), a protease inhibitortablet, 5% (w/v) insoluble polyvinylpyrrolidone and0.05% (v/v) Triton X-100. After centrifugation for 5 minat 14 000 g, the enzyme activity in the supernatantwas measured with an UV-2550 spectrophotometer(Shimadzu, Kyoto, Japan). The initial activity of NADP-malate dehydrogenase (EC 1.1.1.82, NADP-MDH)was measured according to Dutilleul et al. (2003).Assays were performed in 40 mM Tricine-KOH (pH8.3), 150 mM KCl, 1 mM Na2-EDTA, 5 mM DTT,0.2 mM NADPH and 2 mM OAA (oxaloacetate) and

the sample preparation. The activity of superoxidedismutase (EC 1.15.1.1, SOD) was measured accordingto Giannopolitis and Ries (1977). The activities ofascorbate peroxidase (EC 1.11.1.11, APX) and catalase(EC 1.11.1.6, CAT) were measured according to themethod of Mishra et al. (1993).

Measurements of chlorophyll afluorescence transient

Chlorophyll a fluorescence (OJIP) transients beforeand after the high light treatment were measuredwith a Handy PEA fluorometer (Hansatech Instruments,Norfolk, UK). The transients were induced by red lightmeasuring approximately 3000 μmol m−2 s−1 providedby an array of three light-emitting diodes (LEDs, peak650 nm). All measurements were performed with 15-mindark-adapted leaf discs at room temperature.

OJIP transients were analyzed by the JIP-test (Appen-roth et al. 2001, Oukarroum et al. 2009, Strasser andStrasser 1995, Strasser et al. 2000, Yusuf et al. 2010),using the following original data: (1) the fluorescenceintensity at 20 μs, considered to be Fo when all PSIIreaction centers were open; (2) the maximal fluores-cence intensity, Fm (FP, at about 300 ms), assuming thatthe excitation intensity was high enough to close all ofthe PSII RCs; and (3) the fluorescence intensities at 2 ms(J-step, FJ) and 30 ms (I-step, FI). The following param-eters were used for the quantification of PSII behaviorreferring to time 0:

1. The chlorophyll a fluorescence transient from I toP after double normalization between the Fo andFI points, WOI = (Ft− Fo)/(FI− Fo), where Ft is thefluorescence intensity at any time.

2. The relative variable fluorescence at J-step (VJ),VJ = (FJ− Fo)/(Fm− Fo).

3. The specific energy fluxes (per RC) for absorption(ABS/RC), trapping (TR0/RC) and electron transport(ET0/RC).

4. The flux ratios or yields, i.e. the maximalquantum yield of primary photochemistry (φPo =TR0/ABS = Fv/Fm), the efficiency with which atrapped exciton can move an electron into theelectron transport chain further than Q−

A (theexcitation efficiency for electron transport beyondQ−

A , �0 = ET0/TR0) and the quantum yield ofelectron transport (φEo = ET0/ABS).

Measurements of modulated chlorophyllfluorescence parameters

The modulated chlorophyll fluorescence parameters�PSII (PSII actual photochemical efficiency) and qP

398 Physiol. Plant. 143, 2011

(photochemical quenching coefficient) of intact RumexK-1 chloroplasts were measured with a FMS-2 pulse-modulated fluorometer (Hansatech Instruments, Norfolk,UK) integrated with a Chlorolab-2 oxygen electrode(Hansatech Instruments, Norfolk, UK). The actinic lightat 800 μmol m−2 s−1 was offered by the light source.Following the onset of actinic light irradiation, the FS, F′

mand F′

o were recorded when the chlorophyll fluorescenceattained steady-state levels; �PSII = 1 − FS/F′

m and qP =(F′

m− FS)/(F′m− F′

o) (Chen et al. 2007).

Measurement of P700 redox state

The redox state of P700 was measured according toSchansker et al. (2003) and Kim et al. (2001) using aPEA Senior fluorometer (Hansatech Instruments, Norfolk,UK). Red light, measuring 800 μmol m−2 s−1, was pro-duced by an array of four 650-nm LEDs (peak, 650 nm),and the modulated (33.3 kHz) far-red measuring light hasa wavelength of 820 nm was provided by an OD820LED (Opto Diode Corp., Newbury Park, USA). Uponirradiation with a red pulse (800 μmol m−2 s−1), thetransmission at 820 nm in leaves increases gradually,which is mainly caused by the reduction of P700+.

Histochemical detection of superoxide (O−2 )

and hydrogen peroxide (H2O2)

Leaf discs (1.3 cm2) treated with 0 (control) or 1 mMSHAM or nPG were vacuum-infiltrated with 0.1 mg ml−1

nitrotetrazolium blue chloride (NBT) or 1 mg ml−1 3, 3-diaminobenzidine (DAB, pH 3.8) and incubated underhigh light (800 μmol m−2 s−1) conditions for 1 h. Afterthese treatments, leaf discs were decolorized by immer-sion in boiling ethanol (96%) for 10 min. After cooling,the leaf discs were extracted at room temperaturewith fresh ethanol and photographed (Jabs et al. 1996,Thordal-Christensen et al. 1997, Zeng et al. 2010).

Statistical analysis

Least significant difference (LSD) was used to analyzedifferences between the measurements and Tukey’shonestly significant difference (HSD) was used for themultiple comparison using SPSS 16.

Results

SHAM has been widely used to inhibit the AOX pathway(Bartoli et al. 2005, Padmasree and Raghavendra 1999a,1999b, 1999c). However, to avoid its side effects andensure the sufficient inhibition of the AOX pathway,we compared the effects of SHAM with another AOX

inhibitor, nPG (Yoshida et al. 2006), in this study. Theresults showed that the two different inhibitors havesimilar effects on the inhibition of the AOX pathway(data not shown). In addition, there were no differencesin photosynthetic behaviors between leaves treated withSHAM (Figs 3, 5–7) and those treated with nPG (FigsS2–S5). Therefore, we only discussed the effects of theSHAM treatment in the text, and those caused by the nPGtreatment are presented in the Supporting Information.

Given that some components of the photosyntheticelectron transport chain in chloroplasts are similar tothose in the respiratory chain of mitochondria, weexamined whether the concentrations of SHAM andnPG used in this study had direct effects on the photo-synthetic electron transport chain. The results showedthat the treatments with 1 mM SHAM or 1 mM nPGhad no direct effects on the actual PSII photochemicalefficiencies (�PSII) and photochemical quenching coef-ficients (qP) in intact chloroplasts isolated from RumexK-1 leaves (Figs 1 and S1), suggesting that the concentra-tions of SHAM and nPG used in this study had no directeffects on photosynthetic behaviors. Therefore, all of theeffects caused by the SHAM and nPG treatments wereactually because of the inhibition of the mitochondrialAOX pathway.

To determine a working concentration of the inhibitorfor this study, we examined the effects of the inhibitor onthe capacities of the AOX pathways in Rumex K-1 leavesin the dark or in the high light. After the high light treat-ment, the AOX pathway capacities in the control leaves

Fig. 1. Photosystem II actual photochemical efficiency (�PSII) andphotochemical quenching coefficient (qP) in isolated Rumex K-1chloroplasts in the absence (control) or presence of 1 mM SHAM.Chloroplasts were incubated in darkness for 10 min in the presenceof 0 or 1 mM SHAM and then irradiated with 800 μmol m−2 s−1.Each value was obtained after 10 min of irradiation when fluorescenceintensities had achieved steady levels. Results are the average of threeexperiments, using different chloroplast preparations. Different lettersindicate significant difference between the SHAM treatment and controlat P < 0.05, LSD.

Physiol. Plant. 143, 2011 399

Fig. 2. Capacities of alternative respiration (A) and initial activities ofNADP-MDH (B) in Rumex K-1 leaves treated with 0 (control) or 1 mMSHAM in the dark or high light. Leaf discs (0.5 cm2) were infiltratedwith 0 or 1 mM SHAM in the dark for 2 h and then exposed to highlight at 800 μmol m−2 s−1 for 1 h. Means ± SE of three replicates arepresented. Different letters indicate significant difference among the theSHAM treatment and control in the dark and high light at P < 0.05,Tukey’s HSD.

increased significantly (Fig. 2A). The 1 mM SHAM treat-ment inhibited 59% and 55% of the AOX pathwaycapacities in the leaves in the dark and high light,respectively.

NADP-MDH is the key enzyme of the malate-OAAshuttle that is the major machinery for the transportof excess reducing equivalents from chloroplaststo mitochondria for oxidation by the respiratoryelectron transport chain (Noguchi and Yoshida 2008,Raghavendra and Padmasree 2003, Scheibe 2004).The initial activity of NADP-MDH increased byapproximately 181% after high light treatment inthe control leaves. The SHAM treatment did notalter the initial activity of NADP-MDH in the dark,indicating that the 1 mM SHAM treatment had no directeffect on NADP-MDH. However, the 1 mM SHAMtreatment decreased the initial activity of NADP-MDH byapproximately 31% under high light in Rumex K-1 leaves(Fig. 2B), suggesting that SHAM-treated leaves exportedthe reducing equivalents from chloroplasts at lower rates.

The chlorophyll a transient has become one of themost popular tools in photosynthetic research (Jia et al.2010, Kalachanis and Manetas 2010, Mathur et al. 2011,Oukarroum et al. 2009). The shape of the OJIP transientis very sensitive to environmental stresses. Strasser andStrasser (1995) developed a procedure to quantify theOJIP transient, known as the JIP-test. With this test, itis possible to calculate several phenomenological andbiophysical expressions of the photosynthetic electrontransport chain (Strasser and Strasser 1995, Strasser et al.2000, 2004). The JIP-test is thus a powerful tool for the invivo investigation of photosynthetic behaviors, includingthe energy fluxes of absorption, trapping and electrontransport (Strasser et al. 2000, 2004). In the present study,we used the chlorophyll a fluorescence transient (OJIP)to detect changes in photosynthetic behaviors when theAOX pathway was inhibited. In the dark, the SHAMtreatment did not alter the shape of the OJIP transient(Fig. 3A). However, it decreased the fluorescence inten-sities significantly in both the control and SHAM-treated

Fig. 3. Changes of chlorophyll a fluorescence transient (OJIP transient) in Rumex K-1 leaves treated with 0 (control) or 1 mM SHAM in the dark (A) orin the high light (B). Leaf discs (0.5 cm2) were infiltrated with 0 or 1 mM SHAM in the dark for 2 h and then exposed to high light at 800 μmol m−2

s−1 for 1 h. Each transient represents the average of eight samples.

400 Physiol. Plant. 143, 2011

Fig. 4. (A) Traces of P700 reduction in Rumex K-1 leaves treated with0 (control) or 1 mM SHAM after exposure to high light at 800 μmolm−2 s−1. Leaf discs (0.5 cm2) were infiltrated with 0 or 1 mM SHAM inthe dark for 2 h. Each transient represents the average of five samples.(B) Chlorophyll a fluorescence transient from I to P under high lightafter double normalization between the Fo and FI points. Each transientrepresents the average of eight samples.

leaves under high light, and the extent of decrease wasmuch greater in the SHAM-treated leaves than in thecontrol leaves (Fig. 3B). These results indicate that theinhibition of the AOX pathway significantly influencedphotosynthetic behaviors.

The extent of the reduction of P700 was morepronounced in the SHAM-treated leaves than the controlleaves (Fig. 4A). According to the JIP-test, the maximalamplitude of the OJIP transient in the I–P phase afternormalization between the O and I phases decreasedsignificantly when the AOX pathway was inhibitedby 1 mM SHAM under high light (Fig. 4B), suggestingthat the pool size of the PSI end electron acceptorsdecreased (Kalachanis and Manetas 2010, Yusuf et al.2010). The relative variable fluorescence at J-step (VJ)

represents the subsequent kinetic bottlenecks of the

Fig. 5. Photosynthetic parameters derived from the JIP-test analysis ofchlorophyll a fluorescence transients shown in Fig. 3B. Treatments areas same as those described in Fig. 2, where each point represents theaverage of eight samples. All parameters in the leaves of the controlwere taken as 1, whereas those in the leaves treated with 1 mM SHAMwere taken as a percentage of the control. The original value of VJ, �0,φEo, ABS/RC, TR0/RC and ET0/RC in the control leaves were 0.52 ± 0.03,0.48 ± 0.03, 0.36 ± 0.04, 2.53 ± 0.05, 1.87 ± 0.01 and 0.90 ± 0.09,respectively.

electron transport chain, resulting in the momentarymaximum accumulations of Q−

A (Strasser and Strasser1995, Li et al. 2009). In this study, the SHAM treatmentincreased the VJ significantly compared with the controlleaves after high light treatment (Fig. 5). Additionally,compared with the control leaves, the SHAM-treatmentincreased the light absorption per active reaction center(ABS/RC, the average antenna size) and trapping ofexcitation energy per active reaction center (TR0/RC),and it decreased the efficiency of electron transport(�0), the quantum yield of electron transport (φEo) andelectron transport per active reaction center (ET0/RC)after the high light treatment (Fig. 5), suggesting that thebalance between photosynthetic light absorption andenergy utilization was disturbed when the AOX pathwaywas inhibited.

To determine the oxidative stress that the RumexK-1 leaves suffered when the AOX pathway wasinhibited under high light, we detected the in situaccumulation of O−

2 and H2O2 using the NBT andDAB staining procedures, respectively. They showedthat the inhibition of the AOX pathway enhanced theaccumulation of ROS (O−

2 and H2O2) significantly underhigh light in Rumex K-1 leaves (Fig. 6A). The SHAM

Physiol. Plant. 143, 2011 401

Fig. 6. Production of O2− and H2O2 (A) and activities of SOD, APX

and CAT (B) in Rumex K-1 leaves treated with 0 (control) or 1 mMSHAM under high light. Leaf discs (1.3 cm2) treated with 0 (control)or 1 mM SHAM were vacuum- infiltrated with 0.1 mg ml−1 NBT or1 mg ml−1 DAB (pH 3.8) and incubated under high light (800 μmol m−2

s−1) conditions for 1 h. Representative images from five independentexperiments are shown in the figure. Means ± SE of three replicates arepresented. Different letters indicate significant difference between theSHAM treatment and control at P < 0.05, LSD.

treatment also increased the activities of SOD, APX andCAT under high light (Fig. 6B).

Furthermore, compared with control leaves, theSHAM treatment significantly accelerated the decreasein the maximum quantum yield for primary photochem-istry (φPo), the excitation efficiency for electron transportbeyond QA (�0) and the quantum yield for electrontransport (φEo) under high light in the absence of thechloroplast D1 protein synthesis inhibitor chlorampheni-col (CM) in Rumex K-1 leaves (Fig. 7A, C, E). However,in the presence of CM, the SHAM treatment had nosignificant effects on the levels of φPo, �0 and φEo underhigh light (Fig. 7B, D, F).

Discussion

Under high light, excess reducing equivalents generatedin chloroplasts can be transported to other organellesthrough the malate-OAA shuttle and oxidized inmetabolic pathways (Noguchi and Yoshida 2008,Raghavendra and Padmasree 2003, Scheibe 2004).NADP-MDH is the key enzyme in the malate-OAAshuttle (Noguchi and Yoshida 2008, Scheibe 2004). Itis noteworthy that the initial activity of NADP-MDH

noticeably increased in Rumex K-1 leaves under highlight (Fig. 2B), indicating that the malate-OAA shuttlewas activated to transport excess reducing equivalentsgenerated in the chloroplasts to mitochondria. Theequivalents transported from the chloroplasts can beoxidized by the respiratory electron transport chain(Noguchi and Yoshida 2008, Yoshida et al. 2007). Itwas suggested that, in this situation, the mitochondrialnon-phosphorylating electron transport pathway, theAOX pathway, would play a role in the dissipation ofchloroplast-derived reducing equivalents (Dinakar et al.2010b, Yoshida et al. 2007, 2008). The fact that theAOX pathway capacity in the leaves of Rumex K-1significantly increased (Fig. 2A) under high light supportthis suggestion.

The fact that the treatments of intact chloroplastsisolated from Rumex K-1 leaves with 1 mM SHAM andnPG had no effects on �PSII and qP (Figs 1 and S1),together with the fact that neither the SHAM nor thenPG treatments altered the OJIP transients in the dark(Figs 3A and S2A), demonstrate that the 1 mM SHAMand nPG used in this study had no direct effects onphotosynthetic behaviors, which was also demonstratedby Padmasree and Raghavendra (1999a), Bartoli et al.(2005) and Yoshida et al. (2006) in their studies, usingother plant materials. Therefore, the observed effects ofthe SHAM and nPG treatments on the photosyntheticapparatus under high light in this study were becauseof the inhibition of the mitochondrial AOX pathway.It was reported that when the reducing equivalentsaccumulated in the mitochondria, the activity of themalate-OAA shuttle would be reduced by a feed-back mechanism (Padmasree and Raghavendra 2001).Thus, it is reasonable to consider that, under highlight, the significant decrease in the initial activityof NADP-MDH caused by the inhibition of the AOXpathway (Fig. 2B) was because of the accumulation ofreducing equivalents in the mitochondria because theconsumption of reducing equivalents through the AOXpathway was restricted. The significant decrease in theinitial activity of NADP-MDH would inevitably limit theexport of reducing equivalents from the chloroplasts tomitochondria, resulting in the accumulation of reducingequivalents in the chloroplast stroma.

Furthermore, the SHAM and nPG treatments markedlyaltered the OJIP transient under high light (Figs 3Band S2B), suggesting that the inhibition of the AOXpathway significantly influenced the performance of thephotosynthetic machinery. The observation that P700was more reduced in SHAM-treated leaves under highlight (Fig. 4A) indicates that the inhibition of the AOXpathway caused an over-reduction of the PSI acceptorside under high light because of the accumulation of

402 Physiol. Plant. 143, 2011

Fig. 7. Photosynthetic parameters derived from the JIP-test analysis of chlorophyll a fluorescence transients in Rumex K-1 leaves treated with 0(control) or 1 mM SHAM in the absence (A, C, E) or presence (B, D, F) of 1 mM CM. Leaf discs (0.5 cm2) treated with 0 (control) or 1 mM SHAMwere vacuum-infiltrated with 0 or 1 mM CM in the dark and then exposed to high light at 800 μmol m−2 s−1 for 1, 2 and 3 h. Means ± SE of eightreplicates are presented. Different letters indicate significant difference between the SHAM treatment and control at P < 0.05, LSD.

excess reducing equivalents in the chloroplasts. The factthat the maximal amplitude in the I–P phase of theOJIP transient decreased obviously in the SHAM-treatedleaves under high light (Fig. 4B) indicates a decreasein the pool size of the PSI end electron acceptors(Kalachanis and Manetas 2010, Yusuf et al. 2010),further supporting the above suggestion. Moreover, thefact that the relative variable fluorescence at J-step (VJ)

increased significantly in both the SHAM- and nPG-treated leaves compared with the control leaves (Figs 5and S3) suggests that the ratio of Q−

A /QA increased(Li et al. 2009, Strasser and Strasser 1995) under highlight when the AOX pathway was inhibited. Thus, theinhibition of the AOX pathway decreased the electrontransport capacity at the acceptor side of PSII, resultingin an over-reduction of the PSII acceptor side. Inaddition, the electron transport per active reaction center

(ET0/RC) decreased, but the light absorption per activereaction center (ABS/RC) and trapping of excitationenergy per active reaction center (TR0/RC) increased,and the excitation efficiency for electron transportbeyond QA (�0) and the quantum yield for electrontransport (φEo) decreased significantly in the SHAM- andnPG-treated leaves (Figs 5 and S3), indicating that thebalance between photosynthetic light absorption andenergy utilization was disturbed by the over-reductionof the PSII acceptor side because of the inhibition of theAOX pathway under high light. The imbalance betweenphotosynthetic light absorption and energy utilizationwill inevitably lead to the over-excitation of the PSIIreaction centers (Vass 2011).

Under photoinhibitory condition, there are two majorsites of ROS generation in chloroplasts: the end ofphotosynthetic electron transport chain (acceptor side

Physiol. Plant. 143, 2011 403

Fig. 8. Schemes of the possible phenomena in the cell when AOX is inhibited under high light. AOX, alternative oxidase; OAA, oxaloacetate; Pre-D1,precursor to D1 protein; PSII/I, photosystem II/I; ROS, reactive oxygen species.

of PSI) and the PSII reaction centers (Bartoli et al. 2005,Niyogi 1999, 2000, Vass 2011). An over-reduction ofthe PSI acceptor side and over-excitation of PSII reactioncenters because of the inhibition of the AOX pathwaywould inevitably enhance the generation of ROS underhigh light. Our observation that the accumulation of ROS(O−

2 and H2O2) in leaves was increased by the SHAMand nPG treatments (Figs 6A and S4) supports the abovesuggestion. The observation that the activity of the ROSscavenging system (SOD, APX and CAT) increased inSHAM-treated leaves under high light (Fig. 6B) indicatesthat the over-accumulation of ROS was not caused by theinfluence of the SHAM treatment on the activity of ROSscavenging system directly but by the over-reduction ofthe photosynthetic electron transport chain.

The SHAM and nPG treatments caused more severephotoinhibition as they accelerated decreases in φPo

(Figs 7A and S5A), �0 (Figs 7C and S5C) and φEo(Figs 7Eand S5E). How did the inhibition of the AOXpathway accelerate photoinhibition? Some studies havedemonstrated that the accumulation of ROS induced byexcess light energy do not accelerate photodamage toPSII directly but inhibits the repair of the photodamagedPSII at the step of the de novo synthesis of the D1 protein(Murata et al. 2007, Nishiyama et al. 2006, Takahashiand Murata 2008). However, Krieger-Liszkay et al.(2011) recently reported that O−

2 generated in PSI candamage PSII directly. Therefore, the over-accumulationof ROS due to the inhibition of the AOX pathway under

high light (Figs 6A and S4) either damaged PSII directlyor inhibited the repair of the photodamaged PSII. Todistinguish the effect of the inhibition of the AOXpathway on the photodamage of PSII and the inhibitionof the repair of the photodamaged PSII, we used CM,an inhibitor of D1 protein synthesis (Takahashi et al.2009), to inhibit the repair of the photodamaged PSII inthis study. We observed that SHAM and nPG treatmentsdid not accelerate photoinhibition in the presence ofCM, indicated by decreases in φPo (Figs 7B and S5B),�0 (Figs 7D and S5D) and φEo(Figs 7F and S5F), but thetreatments did accelerate photoinhibition in the absenceof CM, indicating that the AOX pathway protected plantleaves against photoinhibition, mainly by alleviating theinhibition of the repair of the photodamaged PSII ratherthan by preventing photodamage to PSII directly.

In conclusion, the inhibition of the AOX pathwayresulted in the loss of a sink of excess reducingequivalents and accelerated the accumulation ofNADPH in chloroplasts, leading to the over-reductionof the PSI acceptor side and over-excitation of the PSIIreaction centers to induce the over-production of ROSin chloroplasts under high light. The over-productionof ROS inhibited the repair of the photodamaged PSII(Fig. 8). Therefore, the mitochondrial AOX pathwayplays an important role in the protection of plantsagainst photoinhibition by alleviating the inhibition ofthe repair of the photodamaged PSII through preventingthe formation of ROS.

404 Physiol. Plant. 143, 2011

Acknowledgements – This work was supported by theState Key Basic Research and Development Plan of China(2009CB118500) and China National Nature ScienceFoundation (30671451 and 30571125).

References

Appenroth KJ, Stockel J, Srivastava A, Strasser RJ (2001)Multiple effects of chromate on the photosyntheticapparatus of Spirodela polyrhiza as probed by OJIPchlorophyll a fluorescence measurements. EnvironPollut 115: 49–64

Aro EM, Suorsa M, Rokka A, Allahverdiyeva Y,Paakkarinen V, Saleem A, Battchikova N, Rintamaki E(2005) Dynamics of photosystem II: a proteomicapproach to thylakoid protein complexes. J Exp Bot 56:347–356

Bartoli CG, Gomez F, Gergoff G, Guiamet JJ, Puntarulo S(2005) Up-regulation of the mitochondrial alternativeoxidase pathway enhances photosynthetic electrontransport under drought conditions. J Exp Bot 53:1269–1276

Chen HX, Li PM, Gao HY (2007) Alleviation ofphotoinhibition by calcium supplement in salt-treatedRumex leaves. Physiol Plant 129: 386–396

Dinakar C, Abhaypratap V, Yearla SR, Raghavendra AS,Padmasree K (2010a) Importance of ROS andantioxidant system during the beneficial interactions ofmitochondrial metabolism with photosynthetic carbonassimilation. Planta 231: 461–474

Dinakar C, Raghavendra AS, Padmasree K (2010b)Importance of AOX pathway in optimizingphotosynthesis under high light stress: role ofpyruvate and malate in activating AOX. Physiol Plant139: 13–26

Dutilleul C, Driscoll S, Cornic G, Paepe RD, Foyer CH,Noctor G (2003) Functional mitochondrial complex I isrequired by tobacco leaves for optimal photosyntheticperformance in photorespiratory conditions and duringtransients. Plant Physiol 131: 264–275

Florez-Sarasa I, Flexas J, Rasmusson AG, Umbach AL,Siedow JN, Ribas-Carbo M (2011) In vivo cytochromeand alternative pathway respiration in leaves ofArabidopsis thaliana plants with altered alternativeoxidase under different light conditions. Plant CellEnviron 34: 1373–1383

Giannopolitis CN, Ries SK (1977) Superoxide dismutases I.occurrence in higher plants. Plant Physiol 59: 309–314

Giraud E, Ho LHM, Clifton R, Carroll A, Estavillo G,Tan Y-F, Howell KA, Ivanova A, Pogson BJ, Millar AH,Whelan J (2008) The absence of alternative oxidase1a inArabidopsis results in acute sensitivity to combined lightand drought stress. Plant Physiol 147: 595–610

Hakala M, Tuominen I, Keranen M, Tyystjarvi T, TyystjarviE (2005) Evidence for the role of the oxygen-evolving

manganese complex in photoinhibition of photosystemII. Biochim Biophys Acta 1706: 68–80

Jabs T, Dietrich RA, Dangl JL (1996) Initiation of runawaycell death in an Arabidopsis mutant by extracellularsuperoxide. Science 27: 1853–1856

Jia YJ, Cheng DD, Wang WB, Gao HY, Liu AX, Li XM,Meng QW (2010) Different enhancement of senescenceinduced by metabolic products of Alternaria alternata intobacco leaves of different ages. Physiol Plant 138:164–175

Kalachanis D, Manetas Y (2010) Analysis of fastchlorophyll fluorescence rise (O-K-J-I-P) curves in greenfruits indicates electron flow limitations at the donorside of PSII and the acceptor sides of both photosystems.Physiol Plant 139: 313–323

Kasahara M, Kagawa T, Oikawa K, Suetsugu N, Miyao M,Wada M (2002) Chloroplast avoidance movementreduces photodamage in plants. Nature 420: 829–832

Kim S-J, Lee C-H, Hope AB, Chow WS (2001) Inhibition ofphotosystems I and II and enhanced back flow ofphotosystem I electrons in cucumber leaf discs chilled inthe light. Plant Cell Physiol 42: 842–848

Krieger-Liszkay A, Kos PB, Hideg E (2011) Superoxideanion radicals generated by methylviologen inphotosystem I damage photosystem II. Physiol Plant142: 17–25

Li PM, Cheng LL, Gao HY, Jiang CD, Peng T (2009)Heterogeneous behavior of PSII in soybean (Glycinemax) leaves with identical PSII photochemistryefficiency under different high temperature treatments. JPlant Physiol 166: 1607–1615

Mathur S, Jajoo A, Mehta P, Bharti S (2011) Analysis ofelevated temperature-induced inhibition of photosystemII using chlorophyll a fluorescence inductionkinetics in wheat leaves (Triticum aestivum). Plant Biol13: 1–6

Mattoo AK, Edelman M (1987) Intramembranetranslocation and posttranslational palmitoylation of thechloroplast 32-kDa herbicide-binding protein. Proc NatlAcad Sci USA 84: 1497–1501

Mishra NP, Mishra RK, Singhal GS (1993) Changes in theactivities of anti-oxidant enzymes during exposure ofintact wheat leaves to strong visible light at differenttemperatures in the presence of protein synthesisinhibitors. Plant Physiol 102: 903–910

Murata N, Takahashi S, Nishiyama Y, Allakhverdiev SI(2007) Photoinhibition of photosystem II underenvironmental stress. Biochim Biophys Acta 1767:414–421

Nishiyama Y, Allakhverdiev SI, Murata N (2006) A newparadigm for the action of reactive oxygen species in thephotoinhibition of photosystem II. Biochim BiophysActa 1757: 742–749

Nishiyama Y, Allakhverdiev SI, Murata N (2011) Proteinsynthesis is the primary target of reactive oxygen species

Physiol. Plant. 143, 2011 405

in the photoinhibition of photosystem II. Physiol Plant142: 35–46

Nishiyama Y, Yamamoto H, Allakhverdiev SI, Inaba M,Yokota A, Murata N (2001) Oxidative stress inhibits therepair of photodamage to the photosynthetic machinery.EMBO J 20: 5587–5594

Niyogi KK (1999) Photoprotection revisited: genetic andmolecular approaches. Annu Rev Plant Physiol PlantMol Biol 50: 333–359

Niyogi KK (2000) Safety valves for photosynthesis. CurrOpin Plant Biol 3: 455–460

Noguchi K, Yoshida K (2008) Interaction betweenphotosynthesis and respiration in illuminated leaves.Mitochondrion 8: 87–99

Ohnishi N, Allakhverdiev SI, Takahashi S, Higashi S,Watanabe M, Nishiyama Y, Murata N (2005) Two-stepmechanism of photodamage to photosystem II: step oneoccurs at the oxygen-evolving complex and step twooccurs at the photochemical reaction center.Biochemistry 44: 8494–8499

Oukarroum A, Schansker G, Strasser RJ (2009) Droughtstress effects on photosystem I content and photosystemII thermotolerance analyzed using Chl a fluorescencekinetics in barley varieties differing in their droughttolerance. Physiol Plant 137: 188–199

Padmasree K, Raghavendra AS (1999a) Importance ofoxidative electron transport over oxidativephosphorylation in optimizing photosynthesis inmesophyll protoplasts of pea (Pisum sativum L). PhysiolPlant 105: 546–553

Padmasree K, Raghavendra AS (1999b) Prolongation ofphotosynthetic induction as a consequence ofinterference with mitochondrial oxidative metabolism inmesophyll protoplasts of the pea (Pisum sativum L).Plant Sci 142: 29–36

Padmasree K, Raghavendra AS (1999c) Response ofphotosynthetic carbon assimilation in mesophyllprotoplasts to restriction on mitochondrial oxidativemetabolism: metabolites related to the redoxstatus and sucrose biosynthesis. Photosynth Res 62:231–239

Padmasree K, Raghavendra AS (2001) Consequence ofrestricted mitochondrial oxidative metabolism onphotosynthetic carbon assimilation in mesophyllprotoplasts: decrease in light activation of fourchloroplastic enzymes. Physiol Plant 112: 582–588

Raghavendra AS, Padmasree K (2003) Beneficialinteractions of mitochondrial metabolism withphotosynthetic carbon assimilation. Trends Plant Sci 8:546–553

Rasmusson AG, Fernie AR, van Dongen JT (2009)Alternative oxidase: a defence against metabolicfluctuations? Physiol Plant 137: 371–382

Schansker G, Srivastava A, Govindjee, Strasser RJ (2003)Characterization of the 820-nm transmission signal

paralleling the chlorophyll a fluorescence rise (OJIP) inpea leaves. Funct Plant Biol 30: 785–796

Scheibe R (2004) Malate valves to balance cellular energysupply. Physiol Plant 120: 21–26

Strasser BJ, Strasser RJ (1995) Measuring fast fluorescencetransients to address environmental questions: the JIPtest. In: Mathis P (ed) Photosynthesis: From Light toBiosphere. Kluwer Academic Publishers, Dordrecht, pp977–980

Strasser RJ, Srivatava A, Tsimilli-Michael M (2000) Thefluorescence transient as a tool to characterize andscreen photosynthetic samples. In: Yunus M, Pathre U,Mohanty P (eds) Probing Photosynthesis: Mechanism,Regulation and Adaptation. Taylor and Francis, London,pp 445–483

Strasser RJ, Tsimill-Michael M, Srivastava A (2004)Analysis of the chlorophyll a fluorescence transient. In:Papageorgiou GC, Govindjee (eds) Chlorophyll aFluorescence: A Signature of Photosynthesis. Advancesin Photosynthesis and Respiration series. KluwerAcademic Publishers, Dordrecht, pp 321–362

Takahashi S, Bauwe H, Badger M (2007) Impairment of thephotorespiratory pathway accelerates photoinhibition ofphotosystem II by suppression of repair process and notacceleration of damage process in Arabidopsis thaliana.Plant Physiol 144: 487–494

Takahashi S, Milward SE, Fan DY, Chow WS, Badger MR(2009) How does cyclic electron flow alleviatephotoinhibition in Arabidopsis? Plant Physiol 149:1560–1567

Takahashi S, Murata N (2008) How do environmentalstresses accelerate photoinhibition? Trends Plant Sci 13:178–182

Thordal-Christensen H, Zhang Z, Wei Y, Collinge DB(1997) Subcellular localization of H2O2 in plants: H2O2

accumulation in papillae and hypersensitive responseduring the barley-powdery mildew interaction. Plant J11: 1187–1194

Tyystjarvi E (2008) Photoinhibition of photosystem II andphotodamage of the oxygen evolving manganesecluster. Coordin Chem Rev 252: 361–376

Van Aken O, Giraud E, Clifton R, Whelan J (2009)Alternative oxidase: a target and regulator of stressresponses. Physiol Plant 137: 354–361

Vass I (2011) Role of charge recombination processes inphotodamage and photoprotection of the photosystem IIcomplex. Physiol Plant 142: 6–16

Walker DA (1988) The use of the oxygen electrode andfluorescence probes in simple measurements ofphotosynthesis. 2nd Edn. Oxygraphics Limited,Sheffield, pp 1–145

Yoshida K, Terashima I, Noguchi K (2006) Distinct roles ofthe cytochrome pathway and alternative oxidase in leafphotosynthesis. Plant Cell Physiol 47: 22–31

406 Physiol. Plant. 143, 2011

Yoshida K, Terashima I, Noguchi K (2007) Up-regulationof mitochondrial alternative oxidase concomitant withchloroplast over-reduction by excess light. Plant CellPhysiol 48: 606–614

Yoshida K, Watanabe C, Kato Y, Sakamoto W, Noguchi K(2008) Influence of chloroplastic photo-oxidative stresson mitochondrial alternative oxidase capacity andrespiratory properties: a case study with Arabidopsisyellow variegated 2. Plant Cell Physiol 49: 592–603

Yusuf MA, Kumar D, Rajwanshi R, Strasser RJ,Tsimilli-Michael M, Govindjee Sarin NB (2010)Overexpression of γ -tocopherol methyl transferase genein transgenic Brassica juncea plants alleviates abioticstress: physiological and chlorophyll a fluorescencemeasurements. Biochim Biophys Acta 1797:1428–1438

Zeng XQ, Chow WS, Su LJ, Peng XX, Peng CL (2010)Protective effect of supplemental anthocyanins onArabidopsis leaves under high light. Physiol Plant 138:215–225

Zhang DW, Xu F, Zhang ZW, Chen YE, Du JB, Jia SD,Yuan S, Lin HH (2010) Effects of light oncyanide-resistant respiration and alternative oxidasefunction in Arabidopsis seedlings. Plant Cell Environ 33:2121–2131

Supporting Information

Additional Supporting Information may be found in theonline version of this article:

Fig. S1. Photosystem II actual photochemical efficien-cies (�PSII) and photochemical quenching coefficients(qP) in isolated Rumex K-1 chloroplasts in the absence(control) or presence of 1 mM nPG.

Fig. S2. Changes of chlorophyll a fluorescence transientsin Rumex K-1 leaves treated with 0 (control) or 1 mMnPG in the dark (A) or in the high light (B).

Fig. S3. Photosynthetic parameters derived from the JIP-test analysis of chlorophyll a fluorescence transientsshown in Fig. S2B.

Fig. S4. Production of O−2 and H2O2 in Rumex K-1

leaves treated with 0 (control) or 1 mM nPG under highlight.

Fig. S5. Photosynthetic parameters derived from the JIP-test analysis of chlorophyll a fluorescence transients inRumex K-1 leaves treated with 0 (control) or 1 mMnPG in the absence (A, C, E) or presence (B, D, F) of1 mM CM.

Please note: Wiley-Blackwell are not responsible forthe content or functionality of any supporting materialssupplied by the authors. Any queries (other than missingmaterial) should be directed to the corresponding authorfor the article.

Edited by P. Gardestrom

Physiol. Plant. 143, 2011 407