Photochemical &Photobiological Sciences

Dynamic Article Links

Cite this: Photochem. Photobiol. Sci., 2011, 10, 947

www.rsc.org/pps PAPER

Role of salicylic acid in alleviating photochemical damage and autophagic celldeath induction of cadmium stress in Arabidopsis thaliana†

WeiNa Zhanga and WenLi Chen*a,b

Received 14th October 2010, Accepted 26th January 2011DOI: 10.1039/c0pp00305k

As a widespread pollutant in the environment, cadmium (Cd) would be accumulated in leaves andcause phytotoxic effect on plants. Salicylic acid (SA), a natural signal molecule, plays an important rolein eliciting specific responses to biotic and abiotic stresses. In our case, the effect of SA on Cd-inducedphotochemical damage and cell death in Arabidopsis was studied. The results illustrated that Cd couldcause a series of physiological events such as chloroplast structure change (e.g. irregular mesophyll cellas well as ultrastructure change), reactive oxygen species (ROS) production and cell death.Furthermore, chlorophyll fluorescence parameters (F v/Fm, qN and ETR) showed a rapid decrease inwild-type (WT) Arabidopsis after treatment with 50 mM CdCl2, identical with the change in chlorophylldelayed fluorescence (DF) intensity. The changes of these parameters showed the damage of Cd toxicityto photosynthetic apparatus. We found that cell death might be autophagic cell death, which might becaused by Cd toxicity induced oxidative stress just like photosynthetic damage. The NahG plants withlower SA accumulation level showed more sensitivity to Cd toxicity, although they exhibited a decreaseboth in chlorophyll fluorescence parameters and DF intensity. Exogenously SA prevented theCd-induced photochemical efficiency decrease and mitigated Cd toxicity. Additionally, SApretreatment could alleviate Cd-induced ROS overproduction. In conclusion, our results suggested thatSA could prevent Cd-induced photosynthetic damage and cell death, which might be due to theinhibition of ROS overproduction.

Introduction

Cadmium (Cd), a non-essential heavy metal, is a widespreadpollutant in the environment. Since Cd accumulates in leavesto a higher level than in other parts of plants,1 research onthe phytotoxic effect of Cd has mainly focused on the field ofphotosynthesis inhibition and has shown that Cd could interferewith the chlorophyll biosynthesis and degradation.2 Previousstudies have associated Cd-induced photosynthesis decrease withpaired photosystem (PS) II photochemistry.3 According to earlierstudies, as the primary site of photoinhibition in thylakoids,PSII plays an important role in the response to environmentalperturbations and stresses in photosynthesis in higher plants.4

And PSII probably contains common sites for heavy metal actionin plants at the oxidizing or reducing side of PSII.5 With regardto the target site of Cd to PSII, it is generally accepted that

aMOE Key Laboratory of Laser Life Science and Institute of Laser Life Sci-ence, South China Normal University, Guangzhou, 510631, China. E-mail:chenwl@scnu.edu.cn ; Fax: +86-20-85216052; Tel: +86-20-85211375-8221;Web: http://sky.scnu.edu.cn/teachers/20085101010697487.htmbCollege of Life Science, Guangdong Key Lab of Biotechnology for PlantDevelopment, South China Normal University, Guangzhou, 510631, China† Electronic supplementary information (ESI) available. See DOI:10.1039/c0pp00305k

the water-oxidizing system of PSII is affected by replacing Mn2+

with Cd2+, thereby inhibiting the reaction of PSII.6 Furthermore,Cd toxicity has been linked to the production of reactive oxygenspecies (ROS), including superoxide ions (O2

-), hydroxyl radicals(HO∑), hydrogen peroxide (H2O2), and so on.7 ROS are highlytoxic and rapidly detoxified by various cellular enzymatic andnon-enzymatic mechanisms. ROS accumulation has been detectedduring various type of cell death process such as autophagic celldeath. Previous studies had shown that rapid accumulation andoverproduction of ROS was the early response of plant to heavymetals.8,9 However, the role of exogenous salicylic acid (SA) onphotochemical damage and autophagic cell death under Cd stressis still unclear and worth further investigation.

SA a natural signal molecule, can elicit specific responses tobiotic and abiotic stresses.9 Studies have shown that SA playedimportant roles in provoking plant resistance to various abioticstresses. For example, exogenous SA could enhance the toleranceof plants to salt, osmotic, drought, chilling and heat stresses.10–13

SA is also known to be involved in plant protection againstheavy-metal stress. Exogenous SA diminished Pb- and Hg-inducedmembranes in rice14 and Cd toxicity in barley,15 maize16 andsoybean plants.17 The protection role of SA mainly includesregulation of normal metabolism, regulation of physiologicalfunction and ROS production.15,17 Apparently, SA has a broad but

This journal is © The Royal Society of Chemistry and Owner Societies 2011 Photochem. Photobiol. Sci., 2011, 10, 947–955 | 947

Publ

ishe

d on

18

Febr

uary

201

1. D

ownl

oade

d by

Lom

onos

ov M

osco

w S

tate

Uni

vers

ity o

n 20

/09/

2013

12:

46:1

5.

View Article Online / Journal Homepage / Table of Contents for this issue

divergent effect on stress acclimation and damage development inplants.

Autophagy is a major process responsible for degradationof cellular proteins and organelles in the vacuole, and it canhave a strong influence on the plant’s response.18 Autophagystabilizes the intracellular environment and maintains cell survivalvia balancing anabolism, metabolism and catabolism. However,overactive autophagy may induce type II programmed cell death,which is called autophagic cell death.19

In this paper, we used an Arabidopsis transgenic line expressingsalicylate hydroxylase gene (nahG) to reduce SA level to study theeffect of exogenous SA on Cd-induced Arabidopsis photosyntheticdamage. Our results detected photosynthetic performance andcell viability in the whole plants and protoplast at differenttimes, due to their different response to Cd and SA. And inprotoplasts we could clearly observe the accumulation processof ROS and the process of cell death using a microscope. Basedon the above results, we supposed that SA application might causepartial protection against heavy metal toxicity in Arabidopsis. Thebeneficial role of SA on plants exposed to Cd appears to bealleviation of photosynthetic damage and autophagic cell deathby regulating ROS production.

Materials and methods

Plant material and experimental design

Arabidopsis (ecotype Columbia 0, wild type) and NahG Arabidop-sis (a transgenic line expressing the salicylate hydroxylase gene(nahG) to reduce SA level) were grown at 22 ◦C in soil culture orMurashige & Shoog (MS) medium in a growth chamber (modelE7/2; Conviron, Winnipeg, MB, Canada) with a 16 h photoperiod(120 mmol quanta m-2 s-1) and 82% relative humidity.

For cadmium chloride (CdCl2) treatment and SA application,three groups were designed: control (pretreatment of seeds withH2O), Cd (treatment with Cd at a final concentration of 50 mM),SA+Cd and SA (pretreatment with 0.5 mM SA for 6 h, followedby 50 mM Cd or not). Seeds were surface sterilized in 1% NaClOsolution for 10 min followed by three washes with sterile distilledwater. The sterilized seeds were sown on solid MS medium.

For soil germination, seeds were sown in plates for 3–4 weeks.Then half of the seedlings presoaked with H2O or SA were wateredwith 50 mM Cd (Cd group or SA+Cd group), while the other halfwere watered with H2O (control group or SA group) and harvestedfor analysis with Cd application for the indicated time.

Trypan blue staining

Leaf samples were taken from 3- to 4-week-old plants and thenimmersed in 50 mM Cd for 24 h or not. Trypan blue staining wasperformed as described by Mauch-Mani and Slusarenko (1996).20

Samples were collected and boiled in lactophenol-Trypan bluestaining solution (10 mL of lactic acid, 10 mL of glycerol, 10 gof phenol, 10 mg of Trypan blue, dissolved in 10 mL of distilledwater) for 1–2 min in a water bath (100 ◦C), and then incubated at25 ◦C for 30 min. They were decolorized in chloral hydrate for 24 h.Samples were stored in 50% glycerol and observed by microscope.

Transmission electron microscopy observation

The leaves were cut into transverse section less than 1 mm wideand fixed in 2.5% glutaraldehyde, post-fixed in buffered 1% osmicacid, and embedded in Epon. Ultrathin sections were examinedby electron microscope (TEM, JEM-1010).

Photochemical efficiency measurement

The setup of delayed fluorescence (DF) biosensor system and thetechnical details of the system have been described by Zhanget al. (2007).21 Leaves were placed inside the chambers of theDF biosensor system for 5 min for dark adaptation before lightirradiation. After a 0.2 s illumination, DF emission in a timewindow of 0.26–5.26 s was recorded, and DF intensity wasobtained by the integration of the DF decay dynamics curve andregistered as count per second (cps).

Measurements and imaging of chlorophyll fluorescence parameters

Chlorophyll fluorescence measurements were performed on leavesin a growth cabinet at 24 ◦C to investigate spatio-temporalchanges in photosynthetic parameters, with a special versionof an Imaging-PAM Chlorophyll Fluorometer (Walz, Effeltrich,Germany). Leaves of different treatments (Control, Cd, SA andSA+Cd) for indicated time were adapted for 15–20 min indark before determining the following Chlorophyll fluorescenceparameters: F 0, Fm, F v/Fm, qN, ETR, Y(II), Y(NPQ) and Y(NO).The details of the Imaging-PAM Chlorophyll Fluorometer systemwere the same as the description by Bonfig et al. (2006).22 Afterkinetics and light curves were recorded, areas of interest (AOIs)were defined by red circles for every leaf, over which all pixel valuesfor various fluorescence parameters were averaged.

Protoplast isolation

Protoplast isolation from Arabidopsis plants (3–4 weeks old)was performed according to a modified procedure as describedpreviously.23,24 Briefly, healthy leaves were sliced with a razor bladeinto small leaf strips (0.5–1 mm) and then vacuum-infiltratedwith enzyme solution [0.2–0.4% (w/v) macerozyme R10 (YakultHonsha, Tokyo, Japan), 1–1.5% (w/v) cellulose R10 (YakultHonsha), 20 mM MES (pH 5.7), 0.4 M mannitol, 10 mM CaCl2

and 20 mM KCl] for 15–20 min, and then incubated for 3 hin darkness without shaking. The enzyme solution containingprotoplasts was filtered with a 75-mm-pore-size nylon mesh. Thefiltrate was centrifuged at 100 g for 3 min at 25 ◦C to pellet theprotoplasts in a round-bottomed tube. Finally, the protoplastswere resuspended in the W5 solution [125 mM CaCl2, 154 mMNaCl, 5 mM glucose, 5 mM KCl and 1.5 mM MES/potassiumacetate (pH 5.6)].

ROS monitor

2¢,7¢-Dichlorofluorescein diacetate (H2DCFDA), a fluorescentprobe for intracellular ROS, was used to monitor ROSproduction.25 Protoplasts isolated from A. thaliana were per-formed and then treated with 50 mM Cd for indicate time after SApretreatment (0.5 h). Protoplasts were incubated with H2DCFDAat a final concentration of 5 mM, and then ROS productionwas visualized using Zeiss LSM 510 laser confocal scanning

948 | Photochem. Photobiol. Sci., 2011, 10, 947–955 This journal is © The Royal Society of Chemistry and Owner Societies 2011

Publ

ishe

d on

18

Febr

uary

201

1. D

ownl

oade

d by

Lom

onos

ov M

osco

w S

tate

Uni

vers

ity o

n 20

/09/

2013

12:

46:1

5.

View Article Online

microscope (LCSM, LSM510/ConfoCor2, Carl–Zeiss, Jena, Ger-many). All images were taken with 20¥ water and 100¥ oil-immersion objectives on the Zeiss LSM 510, and analyzed usingZeiss Rel 3.2 image processing software (Zeiss, Germany). Thefluorescence intensity of DCF was also measured by fluorescencespectrometer (LS55; PerkinElmer, Beacons Weld, Bucks, UK) atroom temperature, with an excitation wavelength of 488 nm andan emission wavelength of 525 nm, with slit width at 2.5 nm. Thefluorescence intensity at 525 nm was used to determine the relativeROS production.

Viability assay

The isolated protoplasts were resuspended in W5 solution, andtreated with Cd and SA for indicated time. Then the protoplastswere incubated with 50 mM Fluorescein diacetate (FDA) for 5 min.The cell viability was measured by flow cytometry and LCSM indark at room temperature. Flow cytometry was performed ona FACSCanto II cytofluorimeter (Becton Dickinson, MountainView, CA, USA) with an excitation wavelength of 488 nm.

Autophagic activity measurement

Lyso Tracker Green (LTG) fluorescence, indicative of autophagyactivity26 was visualized by LCSM using excitation wavelength of488 nm and emission filters of BP 505–550 nm plus LP 650 nm toallow detection chlorophyll-derived red fluorescence, respectively.Leaves were vacuum-infiltrated with 1 mM LTG (DND-26,Invitrogen Molecular Probes, Eugene, Oregon, US) at time pointsafter they were pre-incubated with 10 mM 3-methyladenine (3-MA, a inhibitor of autophagy), 0.5 mM SA or 100 units mL-1

CAT (catalase, a H2O2-specific scavenger) and treated with Cd(kept for additional 1 h in dark before visualization).

Statistics

Experiment data are expressed as means ± S.D. Difference amongtreatments were analyzed by one-way ANOVA, taking P < 0.05(*) or P < 0.01 (**) as significant according to Duncan’s multiplerange test. Data are means of four replicates.

Results

The response of plants and protoplasts to Cd toxicity

Exposure of plants to high Cd concentration resulted in a dramaticdecrease in root length and the fresh/dry weight of both shootand root.27–29 Accordingly, the microscopic analysis of wild-typeleaves stained with Trypan blue showed that Cd could result incell death, and more dead cells were observed in Cd-treated leavesthan in control samples (Fig. 1A). Cells without Cd treatmentdid not accumulate punctuate structures, which is the hint ofautophagic activity (Fig. 3A and B). In contrast, cells treated withCd for 24 h accumulated abundant autolysosomal-like structure(Fig. 3C and D), suggesting that Cd-induced cell death waspossibly an autophagic cell death process. At the same time,chloroplasts of Cd-treated mesophyll cells began to turn irregularand the cells showed morphological changes (Fig. 1B and 1C).Similarly, the ultrastructure of chloroplasts in Cd-treated leaveswas disrupted, characterized by a disturbed shape with a wavy

Fig. 1 Effect of 50 mM Cd on wild-type Arabidopsis. (A) Trypanblue-stained leaves were viewed by light microscope to detect cell deathafter Cd treatment for 24 h. Scale bar = 500 mm. (B) Samples were observedusing a LCSM. Background indicates perspective view. Red signal indicateschlorophyll autofluorescence. Scale bar = 5 mm. (C) Changes in chloroplastafter Cd treatment for 2 h. The changes happened between Cd-treatedseedlings and untreated samples. Chloroplast autofluorescence (red) wasexcited at 488 nm and visualized at 650 nm with a long pass filter. 3Dimages of protoplasts were observed by LCSM. Each value was the mean± S.D. of four replicates. Scale bar = 10 mm. (D) Transmission electronmicrographs (TEM). The damage of chloroplast after Cd treatment wasmonitored. Scale bar = 500 nm.

appearance of grana, stroma thylakoids and the intrathylakoidalspace swollen (Fig. 1D). In order to detect the further toxicityeffect of Cd on cells, protoplasts were used for viability tests. FDAstaining revealed that cell viability declined during Cd application(Fig. 2). After Cd treatment for 1 h, a significant (P < 0.01)decline occurred in Cd-treated protoplasts compared to the control(Fig. 2B).

Effect of Cd and SA treatment on plant photosynthesis

The growth inhibition could be mainly due to the effect of Cd onphotosynthesis rate. Previous studies in our group demonstratedthat changes of delayed fluorescence (DF) value could accuratelyreflect the photosynthetic damage under adverse stresses.24 To in-vestigate the effect of exogenous SA on Cd-induced photosyntheticdamage, NahG plant, which has a lower level of SA than wild-typeplant, was used. A significant (P < 0.05) decline in DF intensityoccurred in seedlings treated with Cd for about 2 h comparedto untreated samples. After 8 h, DF intensity was decreased to47.1% and 29.3% in Cd treated group, and to 70.8% and 58.7%in SA pretreatment group, indicating that the alleviatory effectof SA was realized via preventing the decrease of photochemicalefficiency (Fig. 4).

As parallel experiments, changes in photosynthetic activitywere analyzed using imaging-PAM chlorophyll fluorometer. Thedark-adapted PSII yield (F v/Fm) decreased as well as the PSIIyield during illumination (Y(II)), whereas minimal fluorescence(F 0) and nonphotochemical quenching (qN) increased after Cdtreatment. Obviously, Cd induced damage to NahG was muchmore severe than to wild-type plants. The decrease in F v/Fm, Y(II)and the increase in F 0, qN could be readily discerned from false-color images of these measurement (Fig. 5A and 5B).

F 0 and F v/Fm were used to measure the photochemical effi-ciency and PSII activity. Compared to control, F 0 increased by

This journal is © The Royal Society of Chemistry and Owner Societies 2011 Photochem. Photobiol. Sci., 2011, 10, 947–955 | 949

Publ

ishe

d on

18

Febr

uary

201

1. D

ownl

oade

d by

Lom

onos

ov M

osco

w S

tate

Uni

vers

ity o

n 20

/09/

2013

12:

46:1

5.

View Article Online

Fig. 2 Cd-induced cell viability was time dependent. A: Samples were incubated in 50 mM FDA for 5 min after 50 mM Cd treatment at room temperatureand then observed by LCSM. The viable protoplasts were yellow for both red (chloroplast autofluorescence) and green (FDA); nonviable protoplastswere red for only chloroplast autofluorescence. Scale bar = 100 mm. B: Cell viability in Cd-treated protoplasts, measured by flow cytometry staining withFDA. Protoplasts treated with Cd (50 mM final concentration) were taken from 0 to 2 h at 0.5 h intervals.

Fig. 3 Cd-induced cell death was an autophagic death process. A (B) andC (D): LTG staining of autophagosomal-related structures in wild-typeleaves. LTG-derived fluorescence was apparent in mesophyll cell after24 h Cd stress. Green channel images of LTG-derived fluorescence weresuperimposed with red autofluorescence of chloroplasts. A and C: Scalebar = 50 mm. B and D: Scale bar = 20 mm.

0.013 and 0.051 in wild-type and NahG Arabidopsis respectively,but SA pre-treatment resulted in a slight increase (Fig. 5C). Thepotential beneficial effect of SA was to enhance the antioxidantsystem activity, remove excessive ROS and improve the antioxidantcapacity of Arabidopsis under Cd treatment. In contrast, F v/Fm inCd-treated leaves of wild-type and NahG plant decreased by 7.39%and 12.47%, respectively. However, SA pretreatment inhibitedthe decrease of F v/Fm (0.760–0.790) (Fig. 5D). The above dataindicated that exogenous SA could mitigate F v/Fm reduction.

The variation of qN reflected the nonradioactive energy dis-sipation. In our study, qN in Cd group increased by 11.19%and 18.03% in wild-type plant and NahG, respectively. However,SA pretreatment inhibited the increase of qN (~0.47) and NPQ(~0.185) (Fig. 5E). This phenomenon suggested that Cd inhibitionof Calvin cycle activity was much more severe in NahG than inwild-type plant, indicating that SA played a positive role in thisprocess.

For further confirmation, the ETR changes were examined.Cd application decreased ETR in wild-type and NahG plantand SA pretreatment had a positive effect (ESI, Fig. S1†).When photosynthetically active radiation (PAR) value was below40 mmol quanta m-2 s-1, the light response curve of the relativeETR value in treated and untreated groups were similar. Butthe ETR in Cd-treated leaves declined to a low level thatcould be alleviated by SA pretreatment. This result showedthat SA pretreatment inhibited Cd toxicity effect on electrontransfer system and consequently reduced the poisoning effect onArabidopsis.

Photochemical utilization, regulated heat dissipation (a lossprocess serving for protection) and unregulated heat dissipation(a loss process due to PSII inactivity), which were named as thequantum yields Y(II), Y(NPQ) and Y(NO) respectively, couldassess the excitation energy flux in the PSII in three fundamentallydifferent pathways.21 Our results assessed by the imaging-PAMchlorophyll fluorometer showed that the decreases in Y(II) andY(NPQ) were paralleled by the increase in Y(NO), indicatingthat photosynthesis was inhibited in both wild-type and NahGArabidopsis seedlings (ESI, Fig. S2†).

SA alleviates Cd-induced ROS overproduction and autophagic celldeath in Arabidopsis

It has been reported that ROS was a key signaling moleculeunder many exogenous stimuli, but failure to control the excessaccumulation of ROS could lead to oxidative damage to thecellular component and function including the morphology anddynamics of organelles.30 The level of ROS in Cd-treated leaveswas investigated by the fluorescence of dichlorofluorescein (DCF)produced from the non-fluorescent compound H2DCFDA inthe presence of H2O2.31 Results showed that the DCF intensityincreased with prolonged exposure to Cd, and the DCF level wassignificantly higher in Cd group than in control (Fig. 6A and 6B).ROS level in wild-type Arabidopsis was approximately 1.75 and3-fold higher than control group after Cd treatment for 120 min

950 | Photochem. Photobiol. Sci., 2011, 10, 947–955 This journal is © The Royal Society of Chemistry and Owner Societies 2011

Publ

ishe

d on

18

Febr

uary

201

1. D

ownl

oade

d by

Lom

onos

ov M

osco

w S

tate

Uni

vers

ity o

n 20

/09/

2013

12:

46:1

5.

View Article Online

Fig. 4 Effect of SA on DF intensity in wild-type (A) and NahG (B) Arabidopsis leaves in response to 50 mM Cd treatment at different time. Experimentswere performed four times with similar results.

and 275 min, respectively, while that in NahG Arabidopsis was1.75- and 4.6-fold higher. As examined above, ROS exhibited atime-dependent increase in Cd-treated leaves. However, in the SApretreatment group, Cd-induced ROS production was preventedand DCF fluorescence intensity increased only 31% and 57%in wild-type plant, while 57% and 180% in NahG Arabidopsiscompared to control. To clearly observe the accumulation pro-cess of ROS, we examined the ROS production in protoplastsand results reflected that the protoplasts without Cd treatmentshowed no evident increase in ROS production during the wholeassessment period. However, if they were treated with Cd, astrong DCF fluorescence could be detected after 1 h and ROSlevel was higher in NahG than in wild-type protoplasts duringthe time examined period (Fig. 6C and 6D). In SA pretreatmentgroup, ROS production did not dramatically increased, suggestingthat SA could alleviate the overproduction of ROS. We alsofound that SA alleviated Cd-induced autophagic cell death inwild-type plant. In Cd-treated mesophyll cell, a large number ofgreen dots (representing autophagosome-related structure) weredetected (Fig. 7B) compared to the control and SA group (P< 0.01). The Cd-treated mesophyll cell with or without pre-incubation with 10 mM 3-MA, 0.5 mM SA or 100 units mL-1

CAT for about 2 h, 0.5 h and 0.5 h, respectively, was observed byLCSM showing that less of green dots were present (P < 0.01)(Fig. 7G). The results suggested that Cd-induced cell death couldbe inhibited by 3-MA, which further confirmed that this kind ofcell death should be autophagic death. SA and CAT alleviated celldeath by reducing ROS overproduction (Fig. 7C and E). Overall,SA played a positive effect on protecting plant from oxidativestress.

Discussion

In the present study, we analyzed the mechanism of SA’s beneficialeffect on Arabidopsis exposed to Cd. Results showed that presoak-ing Arabidopsis seeds with 500 mM SA for 6 h before exposure toCd had a protective effect on photosynthesis performance and cellviability.

Cd could induce changes in plant metabolism and finally leadto the inhibition of plant growth.32 Studies have shown thatCd was strongly toxic and even fatal to different species ofplant.33 Cell viability assay by Trypan Blue and FDA stainingshowed that Cd treatment caused reduction of cell viabilityboth in leaves and protoplasts (Fig. 1A and 2). Guo et al.(2009) has reported that Cd-induced root cell death might be anadaptation mechanism by which the plant roots grown in Cd-stressed compartments were self-sacrificed to protect the wholeplant from taking up excessive Cd.34 However, in our study, wespeculated that the time and concentration-dependent cell deathinduced by Cd might be autophagic cell death (Fig. 3). Cd couldalso inhibit photosynthesis and induce damage and dysfunctionof chloroplast.35 Yang (1991) showed that Cd could damagechloroplast submicroscopic structure, mainly via disintegratinggrana stacking structure and forming plasmid ball.36 This resultssuggested that the chloroplast’s ability of capturing light energygreatly decreased, thus affecting the role of a series of functionrelated to photosynthesis.37 Moussa and El-Ganal (2010) alsoreported that Cd treatment caused thylakoid swelling in wheat.29

Our results, obtained with the ultrastructure and 3D reconstructedimages, showed that Cd treatment obviously changed the arrange-ment and morphology of chloroplasts which irregularly arrangedand was almost disintegrated with the prolonged Cd treatmentcompared to the control (Fig. 1B, C). This was in agreementwith our previous research.24 For example, chloroplast thylakoidsswelled, the volume increased, the lamellar structure becameunclear or disappeared, and more osmiophilic granules appearedin chloroplasts (Fig. 1D). Chlorophyll fluorescence imaging hasbeen shown to be capable of revealing spatial and temporalchanges during plant development38 and the environmental effecton several aspects of whole plant physiology.39 Our present studyshowed that SA pre-treatment delayed the damage of PSII reactioncentre in absence of Cd (Fig. 5 and Fig. S2†), indicating that SAexerted a protective effect on photosynthetic apparatus.

SA could induce plant resistance to heavy metals, and such SA-mitigated Cd toxicity has been shown in other plants, includingbarley (Hordeum vulgare),15 soybean (Glycine max),17 maize9 andrice.40 But the mechanisms were different in various plants.

This journal is © The Royal Society of Chemistry and Owner Societies 2011 Photochem. Photobiol. Sci., 2011, 10, 947–955 | 951

Publ

ishe

d on

18

Febr

uary

201

1. D

ownl

oade

d by

Lom

onos

ov M

osco

w S

tate

Uni

vers

ity o

n 20

/09/

2013

12:

46:1

5.

View Article Online

Fig. 5 Comparison of SA effect on chlorophyll fluorescence parameters to wild-type (A) and NahG (B) Arabidopsis leaves in response to 50 mM Cdtreatment. Leaves were sprayed with Cd for 6 days in the presence or absence of SA pretreatment. The false colour code depicted at the bottom of eachimage ranged from 0.000 (black) to 1.000 (purple). C–F: Quantitative analyses of changes in various chlorophyll fluorescence parameters induced by50 mM Cd in the absence or presence of SA pretreatment. Each value was the mean ± S.D. of four independent leaves.

Metwally et al. (2003) reported that seedlings pre-treated with0.5 mM SA could alleviate Cd toxicity.15 Guo et al. (2007) alsoshowed that SA pretreatment alleviated Cd-induced inhibition ofrice growth through elevating enzymatic and many non-enzymaticantioxidants, leading to the alleviation of oxidative damage asindicated by the lower level of H2O2.40

ROS are byproducts of the normal metabolism of oxygen, buthigh levels of ROS are highly toxic, bringing damage to DNA,lipids and proteins. Accumulation of ROS has been detected dur-ing various types of cell death including necrosis and autophagiccell death.41 Studies have shown that uncontrolled productionof ROS could damage organelles, including chloroplasts.24 Inour study we found that the benefit effect of SA-pretreatmentto reduce the production of ROS resulting in the alleviation ofCd-induced oxidative damage and enhancement of Cd tolerance

(Fig. 6). Therefore, we concluded that SA might play an importantrole by inhibiting the activities of catalase (CAT) and ascorbateperoxidase (APX) and increasing the activities of superoxidedismutase (SOD), peroxidase (POD), dehydroascorbate reductase(DHAR) and glutathione reductase (GR). The conclusion was inline with previous reports.40,42

In NahG plants, nahG protein could convert SA into catecholto keep a lower level of SA,43 which could weaken the resistance ofplant to heavy metals. Because SA is a direct scavenger of hydroxylradical and an iron-chelating compound, and could inhibit theFenton-reaction-mediated generation and impact of hydroxylradicals.44 SA may directly act as an antioxidant to scavenge ROSthrough activating antioxidant responses.45 Moreover, NahG plantwas likely subjected to a higher excess of excitation energy, whichcould potentially increase the probability of ROS production,

952 | Photochem. Photobiol. Sci., 2011, 10, 947–955 This journal is © The Royal Society of Chemistry and Owner Societies 2011

Publ

ishe

d on

18

Febr

uary

201

1. D

ownl

oade

d by

Lom

onos

ov M

osco

w S

tate

Uni

vers

ity o

n 20

/09/

2013

12:

46:1

5.

View Article Online

Fig. 6 The DCF fluorescence intensity. Leaves of wild-type (A) and NahG (B) were treated with 50 mM Cd with or without SA pretreatment for theindicated time. Data were the mean ± S.D. of four independent replicates. ROS production was induced by 50 mM Cd in protoplasts of wild-type (C) andNahG Arabidopsis (D). Protoplasts were treated with or without Cd for 2 h, stained with H2DCFDA and observed by LCSM. Protoplasts pretreatedwith SA for 30 min in dim light were treated with Cd for 2 h. Scale bar = 10 mm.

resulting in possible membrane damage and deterioration ofmembrane integrity. However, Zawoznik et al. (2007) reported thatwild-type plants exposed to metal showed more accumulation ofH2O2 than NahG plants,46 which was divergent from our results.This difference may be due to various factors such as differenttreatment time, concentration of CdCl2, plant age, plant organsand so on. And it was possible that the endogenous and exogenousSA played different roles in response to oxidative stress.

Autophagy is a protein degradation process in which cellsrecycle cytoplasmic content in response to environmental stress orplant development.47 However, overactive autophagy may inducetype II programmed cell death which is called as autophagic celldeath. The role of autophagy in cell death has been controversialfor a long period. Two groups provided direct evidence in 2004indicating that autophagy could contribute to cell death in certaincontexts.48,49 In our research, autophagosome-related structurewas detected by LTG staining in Cd-induced cell death, whichcould be inhibited by 3-MA, a known inhibitor of autophagy.

This suggested that Cd-induced cell death was autophagic celldeath (Fig. 3 and 7C).

To our knowledge, this is the first investigation on Cd-inducedautophagic cell death via two fluorescence techniques (DF andchlorophyll fluorescence) to monitor the effect of SA on Cdtoxicity and to unravel the stress mechanisms in a rapid, real-timeand non-invasive way. Taken together, the potential significanceof SA for plant growth in heavy-metal-polluted environmentcan be supported by the findings that SA could alleviate Cdtoxicity to photosynthesis performance and cell viability. Basedon the obtained results, we concluded that Cd could induceROS overproduction resulting in cell death and photochemicaldamage. However, SA, as an antioxidant, could scavenge ROSoverproduction to protect plant from photochemical damage andautophagic cell death. Further investigation is needed to clarifythe relationship and difference between organelle dysfunction,signal transduction and gene expression in wild-type and NahGArabidopsis.

This journal is © The Royal Society of Chemistry and Owner Societies 2011 Photochem. Photobiol. Sci., 2011, 10, 947–955 | 953

Publ

ishe

d on

18

Febr

uary

201

1. D

ownl

oade

d by

Lom

onos

ov M

osco

w S

tate

Uni

vers

ity o

n 20

/09/

2013

12:

46:1

5.

View Article Online

Fig. 7 SA could alleviate Cd-induced autophagic cell death by inhibiting ROS production. The leaves of wild-type plants were pretreated with 10 mM3-MA, 0.5 mM SA or 100 units mL-1 CAT for about 2 h, 0.5 h and 0.5 h respectively, then treated with 50 mM Cd for 24 h and observed by LCSM.Autophagosomal-like structures in leaves can be stained by LTG dye. Green channel images of LTG-derived fluorescence were superimposed with redautofluorescence of chloroplasts. A–F: Scale bar = 20 mm. G: Quantification of autophagosome-related structures. Number of autophagosome-relatedstructure per leaf section was counted and average count was determined by four leaves per treatment. Each value was the mean ± S.D. of four independentreplicates.

Acknowledgements

We are grateful to Dr Dong Jingao (Molecular Plant PathologyLab, College of Life Science, Agricultural University of Hebei) forkindly providing transgenic nahG Arabidopsis seeds. This researchwas supported by the National High Technology Research andDevelopment Program of China (863 Program) [grant number2007AA10Z204], the Chinese Science and Technology Founda-tion of Guangdong Province [grant number 2007A020300008-6], the opening project of MOE Key laboratory of Laser LifeScience, South China Normal University, and the Program forChangjiang Scholars and Innovative Research Team in University(IRT0829).

References

1 C. Alice, A. K. Elizabeth and I. S. Julian, An improved grafting tech-nique for mature Arabidopsis plants demonstrates long-distance shoot-to-root transport of phytochelatins in Arabidopsis, Plant Physiol., 2006,141, 108–120.

2 M. K. Joshi and P. Mohanty, Chlorophyll a fluorescence as a probeof heavy metal ion toxicity in plants, in: G. C. Papageorgiou (ed.):Chlorophyll a Fluorescence: Signature of Photosynthesis. Springer,Dordrecht, 2004, pp. 637–661.

3 E. Franco, S. Alessandrelli, J. Masojidek, A. Margonelli and M. T.Giardi, Modulation of D1 protein turnover under cadmium and heatstresses monitored by [35S] methionine incorporation, Plant Sci., 1999,144, 53–61.

4 C. M. Lu, N. W. Qiu, B. S. Wang and J. H. Zhang, Salinity treatmentshows no effects on photosystem II photochemistry, but increases theresistance of photosystem II to heat stress in halophyte Suaeda salsa,J. Exp. Bot., 2003, 54, 851–860.

5 H. Kupper, F. Kupper and M. Spiller, Environmental relevance ofheavy metal substituted chlorophylls using the example of water plants,J. Exp. Bot., 1996, 47, 259–266.

6 T. Baszynski, L. Wajda, M. Krol, D. Wolinska, Z. Krupa and A.Tukendrof, Photosynthetic activities of cadmium treated tomato plants,Plant Physiol., 1980, 98, 365–370.

7 M. C. Romero-Puertas, M. Rodriguez-Serrano, F. J. Corpas, M.Gomez, L. A. Delrıo and L. M. Sandalio, Cadmium-induced subcel-lular accumulation of O2-and H2O2 in pea leaves, Plant, Cell Environ.,2004, 27, 1122–1134.

8 R. Mittle, Oxidative stress, antioxidants and stress tolerance [J], TrendsPlant Sci., 2002, 7, 405–410.

9 A. Krantev, R. Yordanova, T. Janda, G. Szalaib and L. Popova,Treatment with salicylic acid decreases the effect of cadmium onphotosynthesis in maize plants, J. Plant Physiol., 2008, 165, 920–931.

10 O. Borsani, V. Valpuesta and M. A. Botella, Evidence for a role ofsalicylic acid in the oxidative damage generated by NaCl and osmoticstress in Arabidopsis seedlings, Plant Physiol., 2001, 126, 1024–1030.

11 J. F. Dat, C. H. Foyer and I. M. Scott, Changes in salicylic acid andantioxidants during induced thermotolerance in mustard seedlings,Plant Physiol., 1998, 118, 1455–1461.

12 T. Senaratna, D. Touchell, E. Bunns and K. Dixon, Acetyl salicylic acid(aspirin) and salicylic acid induce multiple stress tolerance in bean andtomato plants, Plant Growth Regul., 2000, 30, 157–161.

13 T. G. Janda, I. T. Szalai and E. Padi, Hydroponic treatment withsalicylic acid decrease the effects of chilling injury in maize (Zea maysL.) plant, Planta, 1999, 208, 175–180.

14 A. Mishra and M. A. Chudhuri, Effect of salicylic acid on heavy metal-induced membrane deterioration in rice, Biol. Plant., 1999, 42, 409–415.

15 A. Metwally, I. Finkermeier, M. Georgi and K. J. Dietz, Salicylic acidalleviates the cadmium toxicity in barley seedlings, Plant Physiol., 2003,132, 272–281.

16 M. Pal, G. Szalai, E. Horvath, T. Janda and E. Paldi, Effect of salicylicacid during heavy metal stress, Proc. 7th Hungarian Congress PlantPhysiol., 2002, 46, 119–120.

17 G. Drazic and N. Mihailovic, Modification of cadmium toxicity insoybean seedlings by salicylic acid, Plant Sci., 2005, 168, 511–517.

18 K. Yoshimoto, Y. Takano and Y. Sakai, Autophagy in plants andphytopathogens, FEBS Lett., 2010, 584, 1350–1358.

19 H. E. Tao, H. J. Wang and Y. Z. Tan, The roles of autophagy in cellsurvival and cell death, Prog. Physiol. Sci., 2008, 39, 37–40.

20 B. Mauch-Mani and A. J. Slusarenko, Production of salicylic acidprecursors is a major function of phenylalanine ammonialyase in theresistance of Arabidopsis to Peronospora parasitica, Plant Cell, 1996, 8,203–212.

21 L. R. Zhang, D. Xing and J. S. Wang, A non-invasive and real-timemonitoring of the regulation of photosynthetic metabolism biosensorbased on measurement of delayed fluorescence in vivo, Sensors, 2007,7, 52–66.

22 K. B. Bonfig, U. Schreiber, A. Gabler, T. Roitsch and S. Berger, Infectionwith virulent and avirulent P.syringae strains differentially affectsphotosynthesis and sink metabolism in Arabidopsis leaves, Planta, 2006,225, 1–12.

23 L. R. Zhang, Q. X. Xu, D. Xing, C. J. Gao and H. W. Xiong, Real-time detection of Caspase-3-like protease activation in vivo usingfluorescence resonance energy transfer during plant programmed celldeath induced by ultraviolet C overexposure, Plant Physiol., 2009, 150,1773–1783.

24 Y. H. Bi, W. L. Chen, W. N. Zhang, Q. Zhou, L. J. Yun and D. Xing,Production of reactive oxygen species, impairment of photosynthetic

954 | Photochem. Photobiol. Sci., 2011, 10, 947–955 This journal is © The Royal Society of Chemistry and Owner Societies 2011

Publ

ishe

d on

18

Febr

uary

201

1. D

ownl

oade

d by

Lom

onos

ov M

osco

w S

tate

Uni

vers

ity o

n 20

/09/

2013

12:

46:1

5.

View Article Online

function and dynamic changes in mitochondria are early events incadmium induced cell death in Arabidopsis thaliana, Biol. Cell, 2009,101, 629–643.

25 C. J. Gao, D. Xing, L. L. Li and L. R. Zhang, Implication of reactiveoxygen species and mitochondrial dysfunction in the early stages ofplant programmed cell death induced by ultraviolet-C overexposure,Planta, 2007, 227, 755–767.

26 Y. Moriyasu, M. Hattori, G. Y. Jauh and J. C. Rogers, Alpha tonoplastintrinsic protein is specifically associated with vacuole membraneinvolved in an autophagic process, Plant Cell Physiol., 2003, 44, 795–802.

27 L. P. Popova, L. T. Maslenkova, R. Y. Yordanova, A. P. Ivanova, A. P.Krantev, G. Szalai and T. Janda, Exogenous treatment with salicylicacid attenuates cadmium toxicity in pea seedlings, Plant Physiol.Biochem., 2009, 47, 224–231.

28 L. A. Belkhadi, H. Hediji, Z. Abbes, I. Nouairi, Z. Barhoumi, M.Zarrouk, W. ChaIbi and W. Djebali, Effects of exogenous salicylic acidpre-treatment on cadmium toxicity and leaf lipid content in Linumusitatissimum L, Ecotoxicol. Environ. Saf., 2010, 73, 1004–1011.

29 H. R. Moussa and S. M. El-Gamal, Effect of salicylic acid pretreatmenton cadmium toxicity in wheat, Biol. Plant., 2010, 54, 315–320.

30 K. Apel and H. Hirt, Reactive oxygen species: metabolism, oxidativestress, and signal transduction, Annu. Rev. Plant Biol., 2004, 55, 373–399.

31 W. Jakubowski and G. Bartosz, 2,7-Dichlorofluorescein oxidation andreactive oxygen species: what does it measure?, Cell Biol. Int., 2000, 24,757–760.

32 L. S. di Toppi and R. Gabbrielli, Response to cadmium in higher plants,Environ. Exp. Bot., 1999, 41, 105–130.

33 I. McCarthy, M. C. Romero-Puertas, J. M. Palma, L. M. Sandalio, F.J. Corpas, M. Gomez and L. A. Delrıo, Cadmium induces senescencesymptoms in leaf peroxisomes of pea plants, Plant, Cell Environ., 2001,24, 1065–1073.

34 B. Guo, Y. C. Liang and Y. G. Zhu, Does salicylic acid regulate antiox-idant defense system, cell death, cadmium uptake and partitioning toacquire cadmium tolerance in rice?, J. Plant Physiol., 2009, 166, 20–31.

35 G. Ouzounidou, M. Moustakas and E. P. Eleftheriou, Physiologicaland ultrastructural effects of cadmium on wheat (Triticum aestivum L.)leaves, Arch. Environ. Contam. Toxicol., 1997, 32, 154–160.

36 D. H. Yang, The effects of heavy metals on the structure and functionof photosynthetic membranes in higher plants, Chin. Bull. Bot., 1991,8, 26–29.

37 L. M. Sandalio, H. C. Dalurzo, M. Gomez, M. C. Romero-Puertas andL. A. del Rio, Cadmiun-induced changes in the growth and oxidativemetabolism of pea plants, J. Exp. Bot., 2001, 52, 2115–2126.

38 J. Soukupova, S. Smatanova, L. Nedbal and A. Jegorov, Plant responseto destruxins visualized by imaging of chlorophyll fluorescence, Physiol.Plant., 2003, 118, 399–405.

39 B. Gielen, H. J. De Boeck, C. M. H. M. Lemmens, R. Valcke, I. Nijsand R. Ceulemans, Grassland species will not necessarily benefit fromfuture elevated air temperatures: a chlorophyll fluorescence approachto study autumn physiology, Physiol. Plant., 2005, 125, 52–63.

40 B. Guo, Y. C. Liang, Y. G. Zhu and F. J. Zhao, Role of salicylic acidin alleviating oxidative damage in rice roots (Oryza sativa) subjected tocadmium stress, Environ. Pollut., 2007, 147, 743–749.

41 M. J. Lenardo, C. K. McPhee and L. Yu, Autophagic cell death,Methods Enzymol., 2009, 453, 71–31.

42 Q. H. Shi and Z. J. Zhu, Effects of exogenous salicylic acid onmanganese toxicity, element contents and antioxidantive system incucumber, Environ. Exp. Bot., 2008, 63, 317–326.

43 T. P. Delaney, S. Uknes, B. Vernooij, L. Friedrich, K. Weymann, D.Negrotto, T. Gaffney, M. Gut-Rella, H. Kessmann, E. Ward and J.Ryals, A central role of salicylic acid in plant disease resistance, Science,1994, 266, 1247–1246.

44 T. C. Dinis, V. M. Maderia and L. M. Almeida, Action of phe-nolic derivates (acetaminophen, salicylate, and 5-aminosalicylate) asinhibitors of membrane lipid peroxidation and as peroxyl radicalscavengers, Arch. Biochem. Biophys., 1994, 315, 161–169.

45 L. P. Popova, L. T. Maslenkova, R. Y. Yordanova, A. P. Ivanova, A. P.Krantev, G. Szalai and T. Janda, Exogenous treatment with salicylicacid attenuates cadmium toxicity in pea seedlings, Plant Physiol.Biochem., 2009, 47, 224–231.

46 M. S. Zawoznik, M. D. Groppa, M. L. Tomaro and M. P. Benavides,Endogenous salicylic acid potentiates cadmium-induced oxidativestress in Arabidopsis thaliana, Plant Sci., 2007, 173, 190–197.

47 D. Bassham, Plant autophagy—more than a starvation response, Curr.Opin. Plant Biol., 2007, 10, 587–593.

48 S. Shimizu, T. Kanaseki, N. Mizushima, T. Mizuta, S. Arakawa-Kobayashi, C. B. Thompson and Y. Tsujimoto, Role of Bcl-2 familyproteins in a non-apoptotic programmed cell death dependent onautophagy genes, Nat. Cell Biol., 2004, 6, 1221–1228.

49 L. Yu, A. Alva, H. Su, P. Dutt, E. Freundt, S. Welsh, E. H. Baehreckeand M. J. Lenardo, Regulation of an ATG7-beclin1 program ofautophagic cell death by caspase-8, Science, 2004, 304, 1500–1502.

This journal is © The Royal Society of Chemistry and Owner Societies 2011 Photochem. Photobiol. Sci., 2011, 10, 947–955 | 955

Publ

ishe

d on

18

Febr

uary

201

1. D

ownl

oade

d by

Lom

onos

ov M

osco

w S

tate

Uni

vers

ity o

n 20

/09/

2013

12:

46:1

5.

View Article Online