ARTICLE IN PRESS

Journal of Plant Physiology 162 (2005) 977—984

KEYWORDAllium satCadmium;Catalase;DismutaseOxidativePeroxidaseSuperoxide

0176-1617/$ - sdoi:10.1016/j.

Abbreviationmalondialdehy�CorrespondE-mail addr

www.elsevier.de/jplph

Cadmium accumulation and oxidative burst ingarlic (Allium sativum)

Haiyan Zhang, Yingnan Jiang, Zhenyan He, Mi Ma�

Key Laboratory of Photosynthesis and Molecular Environmental Physiology, Institute of Botany, Chinese Academy ofSciences, Beijing, Xiangshan 100093, PR China

Received 29 July 2004; accepted 18 October 2004

Sivum;

;stress;;

ee front matter & 200jplph.2004.10.001

s: AOS, active oxygende; POD, peroxidase;ing author. Tel.: +8610ess: mami@ibcas.ac.c

SummaryTo investigate the temporal sequence of physiological reactions of garlic (Alliumsativum) to cadmium (Cd) treatment, seedlings developed from cloves were grown inincreasing concentrations of CdCl2, ranging from 1–10mM, for up to 8 days in sand.Analysis of Cd uptake indicated that most Cd accumulated in roots, but some was alsotranslocated and accumulated in leaves at longer exposure time (after 12 h) andhigher concentrations (5 and 10mM) of CdCl2. Changes in activities of antioxidativeenzymes, including superoxide dismutase (SOD), peroxidase (POD) and catalase (CAT),were characterized in leaves of garlic seedlings. Cd (5 and 10mM) initially inhibitedthe activities of SOD and CAT but thereafter recovered or even increased comparedwith control plants. POD activities at 5 and 10mM of Cd increased more than 3–4 timesover control plants within 12 h and then dropped, but were still higher than controls atthe end of the experiment. Otherwise lipid peroxidation enhanced with the increasingof incubation time and concentrations of external Cd. Leaves exposed to 1mM CdCl2showed a less pronounced response and only a small reduction in shoot growth. Theseresults suggested that in leaves of garlic seedlings challenged by CdCl2 at higherconcentrations, induction of these various enzymes is part of a general defensestrategy to cope with overproduction of reactive oxygen. The possible mechanism ofantioxidative enzymes changing before Cd accumulation in leaves of garlic seedlings isdiscussed.& 2005 Elsevier GmbH. All rights reserved.

5 Elsevier GmbH. All rights reserved.

species; AP, ascorbate peroxidase; CAT, catalase; EDTA, ethylene diaminetetraacetic acid; MDA,SOD, superoxide dismutase8259 9422; fax: +8610 6259 0833.n (Mi Ma).

ARTICLE IN PRESS

H. Zhang et al.978

Introduction

Heavy metals are ubiquitous in nature. Heavymetals are emitted from a wide spectrum ofnatural and man-made sources such as volcaniceruption, fossil burning, industrial emissions, auto-mobile exhausts and sewage disposals. Cadmium(Cd) is a major environmental contaminant and wasclassified as an element of ‘intermediate toxicity’by Duxbury (1985). Exposures to high Cd concen-tration have been found to be carcinogenic,mutagenic and teratogenic for a large number ofanimal species (Degraeve, 1981). Recently, it hasbeen reported that low concentrations of Cd(5–10 mg/kg) can mimic the in vivo effects ofestrogen and induce uterine hyperplasia or earlyonset of puberty in animals (Johnson et al., 2003).In plants Cd accumulation causes reductions inphotosynthesis, diminishes water and nutrientuptake and results in visible symptoms of injury inplants such as chlorosis, growth inhibition, brown-ing of root tips, and finally death (Kahle, 1993;Drazkiewicz et al., 2003). Since the presence of Cdprevents the development of a normal vegetationcover, biotechnological efforts are underway todevelop more stress-resistant species. For thispurpose, it is important to understand the mechan-isms of Cd toxicity and tolerance in plants.

Cd was also known to cause the burst of activeoxygen species (AOS) in plant tissues, leading todevelopment of secondary oxidative stress. Tocontrol the level of AOS, plants can induce a seriesof detoxification reaction catalysed by antioxida-tive enzymes. However, contrasting results havebeen reported about the activities of antioxidativeenzymes affected by Cd. In Phaseolus vulgarisseedlings, Cd induced elevated guaoacol peroxi-dase, but decreased catalase activities (Shaw,1995). Severe diminution of ascorbate peroxidase,catalase, and glutathione reductase activities werealso found in Populus canescens roots. (Schutzend-subel et al., 2002). An increase in CAT activityunder Cd stress has been found in other species.(Vitoria et al., 2001). Cd also caused lipid perox-idation suggesting that the tissues suffered fromoxidative stress. The involvement of antioxidativeenzymes in plant responses against Cd toxicity isunclear because Cd does not belong to the groupof transition metals like copper, iron, and zinc,which may induce oxidative stress via Fenton-typereactions.

Garlic is well known for its characteristicresistance to biotic and abiotic environmentalstresses, such as viral, bacterial and oxidativestress. Cd was recognized as causing oxidativestress in plants, so garlic may have a strong

resistance to Cd, but little is known about theeffects of Cd on the physiological processes ofgarlic seedlings. In this study, we investigated thetemporal sequence of physiological reactions,including changes in shoot growth, Cd contents,antioxidative enzymes (SOD, POD, and CAT), lipidperoxidation, and protein occurring in roots orleaves of garlic seedlings after Cd exposure. Wealso discussed the possible mechanisms of thetolerance of garlic seedlings to Cd stress.

Materials and methods

Plant material

Seedlings were developed from cloves of garlic(Allium sativum) in polyethylene pots with 2 kgclean sand. The experiment was carried out undergreenhouse conditions. All pots were watered withdeionized water. After 3 weeks, the plants weretreated with 1, 2, 5, 7 and 10mM CdCl2 solution for8 days and harvested at each sample date foranalysis.

Shoot length measurement and sampling

Shoot lengths were measured before Cd exposureand after 8-day Cd exposure with a centimeterruler. The differences between these two valueswere used to indicate shoot growth.

Samples for analysis of enzymes’ activities,malondialdehyde, and soluble proteins were col-lected after 6, 12, 36, 72 h and 8 days of Cdtreatment. From the top down, the first leaves of8–10 seedlings (for the analysis of enzyme activity)were mixed and 3 aliquots plunged into liquidnitrogen and stored at �70 1C until use. In the sameway, the second leaves (for the analysis ofmalondialdehyde and soluble protein) were har-vested and stored.

Cd determination

Plants were uprooted from the sand carefully,washed thoroughly with running deionized water,divided into leaves and roots, and dried to aconstant weight. Three hundred milligram drysamples were digested with a mixture of HNO3

and HClO4 (10:1, v/v). Cd was determined byInductively Coupled Plasma-atomic Emission Spec-troscopy (ICP-AES) with a standard method.

ARTICLE IN PRESS

Response to Cd in garlic 979

Malondialdehyde and protein determination

For measurement of malondialdehyde (MDA)content, 4mL of 20% trichloroacetic acid contain-ing 0.5% thiobarbituric acid was added to a 1-mLaliquot of the supernatant. The mixture was heatedat 95 1C for 30min and then quickly cooled in an icebath. After the tube was centrifuged at 10,000 g for10min, the absorbance of the supernatant was readat 532 nm. The value for the nonspecific absorptionat 600 nm was subtracted from the 532 nm reading.The concentration of MDA was calculated usingMDA’s extinction coefficient of 155mM�1 cm�1

(Heath and Packer, 1968).Protein was assayed with protein reagent 0.01%

(w/v) Coomassie Brilliant Blue G-250, 4.7% (w/v)ethanol, 8.5% (w/v) phosphoric acid as described byBradford (1976) using bovine serum albumin as astandard.

0

2

4

6

8

10

12

14

0 1 2 5 7 10

CdCl2 concentration (mM)

Sho

ot g

row

th (

cm)

Figure 1. Growth of shoots of garlic seedlings during 8days of CdCl2 treatment. Each value is the mean of 8individual replicates ð�SDÞ:

Enzyme assay

Activities of catalase (CAT; EC 1.11.1.6), perox-idase (POD; EC 1.11.1.7) and superoxide dismutases(SOD; EC 1.15.1.1) were assayed in frozen leaftissue extracts. For enzyme extraction, frozenleaves (E0.5 g) were ground in liquid nitrogen ina prechilled mortar. CAT was assayed with amodified method as described by Barber (1980).The reaction mixture contained 0.5mmol hydrogenperoxide, 0.01mmol potassium phosphate buffer(pH 7.0) and 1mL of suitably digested enzymeextract in a final volume of 10mL. After incubatingit for 5min at 25 1C, the reaction was stopped byadding 5mL 2N H2SO4. The residual H2O2 wastitrated with 0.1 N KMnO4. A blank was maintainedwith the reaction mixture at zero time. One unit ofCAT activity was expressed as micromole H2O2

oxidized mg�1 protein min�1.The POD activity was determined at 25 1C with

guaiacol (Zhang et al., 1995). In the presence ofH2O2, POD catalyses the transformation of guaiacolto tetraguaiacol (brown product). The oxidation ofguaiacol was measured by the increase in absor-bance at 470 nm for 1min ð� ¼ 26:6 Lmmol�1 cm�1Þ:The reaction mixture contained 50 mL of 20mMguaiacol, 2.8mL of 10mM phosphate buffer (pH7.0), and 0.1mL enzyme extract. The reaction wasstarted with 20 mL of 40mM H2O2. One unit of PODactivity was expressed as micromole guaiacoloxidized mg�1 fresh weight min�1.

SOD was extracted with 50mM phosphate buffer(pH 7.8) containing 1% polyvinyl pyrrolidine (Zhanget al., 1995; Beauchamp and Fridovich, 1971). Theactivity of SOD was assayed by measuring its ability

to inhibit to the photochemical reduction ofnitroblue tetrazolium. The reaction mixture(3.0mL) consisted of 50mM phosphate buffer (pH7.8), 13mM methionine, 75 mM nitroblue tetrazo-lium, 0.1mM EDTA, 2 mM riboflavin and 20 mL ofenzyme extract. The reaction mix which was notexposed to light did not develop color and served ascontrol. The absorbance was measured spectro-photometrically at 560 nm. One unit of SOD activitywas defined as the amount of enzyme required tocause 50% inhibition of the rate of nitrobluetetrazolium reduction at 560 nm.

Statistics

In all experiments three replicates were per-formed for each time of exposure to Cd and eachtested concentration. Data presented here are themeans7SD of three independent experiments.

Results

Exposure for 8 days of 3-week-old garlic seedlingsto different concentrations of CdCl2 solution didnot produce any visible toxicity symptoms butcaused obvious retardations of growth under higherCdCl2 concentrations (2, 5, 7, 10mM) (Fig. 1).Exposure to CdCl2 at 1mM caused 3.54% reduction,however at 10mM it caused 22.46% reduction.Whereas there were no significant differences togrowth between 2, 5, 7, 10mM CdCl2 treatment,suggesting garlic had great resistance to Cd stress.

In relation to Cd uptake, concentration in rootswas generally, by one or two orders of magnitude,higher than that in shoots, particularly at higherCdCl2 concentrations tested, with very low translo-cation of Cd to shoots (Fig. 2). Cd accumulation inroots was found to be dose and time-dependent

ARTICLE IN PRESS

0

500

1000

1500

2000

2500

3000

0 6h 12h 36h 72h 8d

Time

0 6h 12h 36h 72h 8d

Time

Cd

cont

ent o

fro

ots

(µg/

gDW

)

Cd

cont

ent o

fle

aves

(µg

/gD

W)

0

5

10

15

20

25

30

Figure 2. Cd content in roots (left) and leaves (right) of garlic seedlings treated with CdCl2 (B, 0mM; &, 1mM; n,5mM; �, 10mM). Each value is the mean of three individual replicates (7SD).

0100200300400500600700800

6h 12h 36h 72h 8d

Time

MD

A (

nmol

/gD

W)

Figure 3. MDA content in leaves of garlic seedlingstreated with CdCl2 (B, 0mM; &, 1mM; n, 5mM; *,10mM). Each value is the mean of three individualreplicates (7SD).

02468

10121416182022

CAT POD SOD

Enz

yme

activ

ity (

IU/m

g pr

otei

n)

Figure 4. Activities of antioxidative enzymes in leaves ofgarlic seedlings. Leaves from control plants wereharvested at the same time when Cd-treated plants wereharvested and were used to determine antioxidativeenzymes ðn ¼ 3persamplingdateÞ: The bars indicate over-all means from all sampling dates (n ¼ 15; 7SD).

H. Zhang et al.980

especially at higher CdCl2 concentrations (5 and10mM). There was no significant variation atdifferent treatment time under 1mM CdCl2 com-pared with higher external Cd concentrations (Fig.2). Cd in leaves did not obviously accumulatewithin 12 h, thereafter, leaves accumulated0.7–10.7 times Cd over controls at 5 and 10mMCdCl2. Cd in roots increased 124% and 439% within6 h, respectively, at 5 and 10mM CdCl2, thereafter,Cd accumulated at higher rate and Cd contents inroots obtained as high as 1100 mg/ g DW and 2954 mg/g DW, respectively, at the end of the experiment.

Exposure to CdCl2 resulted in an accumulation oflipid peroxidation products in leaves (Fig. 3). Thelevels of MDA increased with CdCl2 concentrationand treatment time. Although no significant in-crease in lipid peroxidation within 6 h of stressoccurred at any of the concentrations tested, morerapid increases of MDA under 5 and 10mM CdCl2than that under 1mM CdCl2 were found after thattime. For example, seedlings of 1mM CdCl2 showed26% increase over control plants up to 12 h,followed by slight further change, whereas at 5and 10mM CdCl2 MDA contents rose sharply after

6 h of stress and increased 150% and 300%,respectively, at the end of the experiment.

In leaves of control seedlings, activities ofantioxidative enzymes were determined over anexperimental period of 8 days in parallel withmeasurements in Cd-treated leaves. In this period,enzyme activities of controls varied only slightlyaround their means (Fig. 4). CAT activity of controlseedlings was about 16-fold higher than PODactivity.

SOD activity in leaves significantly increased at1mM CdCl2 during the 6th hour of stress (Fig. 5). Atthe same time, the enzyme activities were inhib-ited by about 30% at 5 and 10mM CdCl2. However,this stimulation/inhibition was only transient. After6 h, treatment with all CdCl2 concentrations testedresulted in a similarity of SOD activities to thosefound in controls and increased significantly at theend of the experiment except for 10mM CdCl2 thatshowed enzyme activity increased 42% at 72 h stressand then decreased 25% after 8-day stress com-pared with control.

ARTICLE IN PRESS

Response to Cd in garlic 981

In contrast to SOD, which was suppressed byhigher concentrations of Cd, POD activities inleaves were initially significantly stimulated by 5and 10mM CdCl2, after 12 h, the activities of thoseplants dropped, but were still obviously higher thanthose of control plants, and POD activities at 10mMdropped more quickly than those at 5mM (Fig. 6).Exposure to 1mM CdCl2 induced slight stimulationof POD activities within 36 h, at later stages, theactivities were strongly increased by nearly fourtimes over controls.

CAT activities were also initially dramaticallysuppressed after exposure to 5 and 10mM CdCl2,after 12 h, the activities had recovered slowly(10mM) to the extent of control or even increasedby 32% (5mM) over the control at the end of theexperiment (Fig. 6). Treatment with 1mM CdCl2caused slight increases in activities of CAT withincubation time of Cd and at the end of theexperiment the increase of CAT activity was very

0

100

200

300

400

500

600

6h 12h 36h 72h 8dTime

PO

D(%

)

Figure 6. Activity of POD (left) and CAT (right) in leaves of ga10mM). The activity was expressed relative to the activity iindividual replicates (7SD).

0

50

100

150

200

6h 12h 36h 72h 8d

Time

SO

D (

%)

Figure 5. Activity of SOD in leaves of garlic seedlingstreated with CdCl2 (B, 1mM; &, 5mM; n, 10mM). Theactivity was expressed relative to the activity in controlplants (100%). Each value is the mean of three individualreplicates (7SD).

similar to that found in leaves at 5mM CdCl2. TheCd-mediated changes in CAT activity were muchsmaller than that of SOD and POD.

The contents of protein in leaves kept nearlyconstant (1mM) or obviously increased (5 and10mM) with increasing Cd supply within 36 h ofCdCl2 exposure, thereafter, these contents recov-ered to the extent of control plants (10mM) oreven decreased lower than controls (1 and 5mM)(Fig. 7).

Discussion

It is well known that Cd inhibits plant growth,this is consistent with the results presented in thisstudy (Fig. 1). Otherwise it is notable that elonga-tion growth of garlic shoot was more resistant to Cdthan that of other species (Fig. 1). For example,only one of four varieties of radish survived theeffects of 4 days of 1mM CdCl2 treatment (Vitoriaet al., 2001). A concentration of 50 mM Cd resultedin more than 50% growth inhibition in poplar shootsafter 48 h (Schutzendubel et al., 2002). Further-

6h 12h 36h 72h 8dTime

0

50

100

150

200

250

300

CA

T(%

)

rlic seedlings treated with CdCl2 (B, 1mM; &, 5mM; n,n control plants (100%). Each value is the mean of three

0

50

100

150

200

250

300

350

6h 12h 36h 72h 8d

Time

Pro

tein

(m

g/gD

W)

01mM5mM10mM

Figure 7. Protein content in leaves of garlic seedlingstreated with CdCl2. Each value is the mean of threeindividual replicates (7SD).

ARTICLE IN PRESS

H. Zhang et al.982

more, small and similar retardation of shoot growthat higher concentrations (2, 5, 7, 10mM) of CdCl2were found in our study. All these results implygarlic possesses strong resistance to Cd.

Cd distribution in plant tissues is an indication ofa change in chemical form resulting from com-plexation with plant-produced ligands. Most of Cdtransported into plants are chelated by chemicalcompounds such as organic acids and phytochela-tins. Only traces of Cd can be detected as a free ionand Cd toxicity is related to the free Cd contam-ination. Therefore in plants which are not hyper-accumulators of Cd, roots have been shownpreviously to be the major site of phytochelatinsynthesis and hence Cd accumulation (Di Cagno etal., 1999; Zhu et al., 1999). We also showed that Cdaccumulated in roots at a faster rate and to a muchhigher concentration than in leaves after Cdapplication (Fig. 2). Similar results have been foundin soybean in which about 98% of the accumulatedCd was retained by roots (Becher and Hofner,1994). Cd in leaves did not obviously accumulate atany time of stress under 1mM CdCl2 and within 12 hunder 5 and 10mM CdCl2 (Fig. 2). These observa-tions suggest that the cellular metal homeostasis inleaves can be maintained in environments with lowCdCl2 concentration or a short period of Cdincubation. This pattern of Cd changes is distinctlydifferent from that found in Scots pine whichaccumulated a majority of Cd during the first 24 hafter 50 mM Cd exposure (Schutzendubel et al.,2001).

A variety of abiotic stresses including heavymetals cause molecular damage to plant cellseither directly or indirectly through the burst ofAOS (Lin and Kao, 2000; Cuypers et al., 2002).Protonation of O2

� can produce the hydroperoxylradical (OH�, H2O2), which can convert fatty acidsto toxic lipid peroxides, destroying biologicalmembranes. MDA formation is used as the generalindicator of the extent of lipid peroxidationresulting from oxidative stress. In the presentstudy, after 8 days of Cd exposure, MDA contentsin leaves increased about 0.5 –3 times (Fig. 3). Thissuggests that Cd indirectly leads to production ofsuperoxide radicals, resulting in increased lipidperoxidative products and oxidative stress in garlicplants. MDA was not greatly affected during most ofthe experimental periods by 1mM CdCl2 treatment.This indicates that the capabilities of garlic seed-lings to adapt to lower concentration of Cd may berelated to a low degree of lipid peroxidation andthus a higher rate of growth was maintained (Figs.3, 1). It is notable that increases of MDA contentoccurred earlier than Cd accumulation in leaves butlater than Cd accumulation in roots. It is possible

that Cd accumulation in roots may set off certaincompounds such as H2O2 and be transferred toleaves to induce a series of physiological reactionsin leaves is.

SOD catalyses the dismutation of O2� to H2O2 and

O2 and plays a key role in quenching active oxygen.The lack or stimulating effects of 1mM CdCl2 stresson SOD suggests that maintenance of physiologicalactivities under low concentration of Cd (Fig. 5), asshown by shoot growth, could be related to thesteady or increased SOD activity to scavenge activeoxygen. The continuous increase in SOD during 72 hat 5mM and 10mM illustrated that AOS accumu-lated with the incubation time of Cd. The decline inSOD activity after prolonged Cd exposure especiallyat 10mM indicated that the scavenging function ofSOD was impaired with prolonged, severe Cd stress(Fig. 5). However, after 8 days of 5mM Cdtreatment, SOD activity in leaves was nearly 51%higher than controls. These observations were inagreement with the results from Alyssum species(Schickler and Caspi, 1999) that had definiteincreases in SOD activity in response to Cd butthese fingdings were different from the reportsfrom polar which exhibited inhibitory effects of Cdon SOD activity (Schutzendubel et al., 2002).

POD catalyses H2O2-dependent oxidation of sub-strate. Garlic seedlings were able to maintain highlevels of POD activity for detoxifying active oxygenin response to higher concentration of Cd (Fig. 6).Other studies have reported increases, decreasesand no changes in POD activity in response to Cdexposure (Shaw, 1995; Schutzendubel et al., 2001,2002). CAT eliminates H2O2 by breaking it downdirectly to form water and oxygen. In the garlicseedlings studied here, CAT activity was suppressedby 5 and 10mM CdCl2 within 36 and 72 h respec-tively and then increased slowly (Fig. 6). Similarpatterns of POD and CAT activity have beenreported in plant tissues under NaCl salinity (Mittaland Dubey, 1991) and Fe toxicity (Hendry andBrocklebank, 1985). CAT is less efficient than PODin H2O2 scavenging because of its low substrateaffinity, so as long as the stress is not too strong forthe plant defense capacity, the main response toheavy metal is an increase in SOD and PODactivities along with a decrease in CAT (Siedleckaand Krupa, 2002). By extending the incubation timeof Cd, a portion of POD could be inactivated by thehigh level of Cd accumulated in leaves, CAT wasactivated or synthesized to compensate for insuffi-ciency of POD. Lidon and Teixeira also observedthat in Mn-treated rice plants, both under Mn-deficiency as well as at Mn excess, there was asignificant increase in SOD and ascorbate perox-idase (AP) activities, while at physiological Mn

ARTICLE IN PRESS

Response to Cd in garlic 983

supply CAT was the predominant oxy-radical sca-venger (Lidon and Teixeira, 2000). This is consistentwith the results obtained from control plants in thepresent study that showed CAT activity was muchhigher than POD (Fig. 4).

After Cd incubation, the activities of these threeenzymes in leaves changed (within 6 h) followed bylipid peroxidation (after 6 h), then Cd accumulation(after 12 h) (Figs. 2, 3, 5 and 6). This finding isimportant because it shows that these earlierreactions were not caused directly by changes inCd content in leaves. Cd content in roots increasedmore than 1–4 times at higher CdCl2 concentrationwithin 6 h. We assumed that Cd built up in roots canresult in a significant accumulation of a substancewhich can act as signaling molecule in the activa-tion of common plant defenses in leaves. H2O2 mayplay an important role in this process because itsformation after Cd exposure has been detected indifferent species and experimental systems such aspotato tuber discs, suspension cultures of tobaccocells and it has been shown to be a signalingmolecule (Stroinski and Zielezinska, 1997; Karpinskiet al., 1999; Piqueras et al., 1999). It has beenobserved that in plant-pathogen interactions, H2O2

induces an orchestrated sequence of reactionsincluding the activation of POD, the stimulation ofsecondary metabolism, structural changes andeventually cell death (Alvarez and Lamb, 1997),and in Scots pine plant, H2O2 triggers a develop-mental program leading to xylogenesis (Schutzen-dubel et al., 2001). It is possible that accumulatedH2O2 as a result of an abrupt increase of Cd in rootsis translocated to shoots to induce lipid peroxida-tion and antioxidative defenses in leaves of garlicseedlings.

Earlier studies have shown that increased con-centration of heavy metal and duration of exposuresignificantly inhibited protein content of Nymphaeaalba (Vajpayee et al., 2000) and duck weed(Tripathi and Smith, 1996). Our results showed that5 and 10mM Cd induced increase of protein in garlicleaves within 72 h and decreased at later stages(Fig. 7). These results suggest that the rate ofbiosynthesis of proteins was higher than that ofdegradation at Cd exposure within 72 h. Cd cancause oxidative stress as we just discussed, and aseries of proteins are accumulated or synthesizedunder oxidative stresses. In response to oxidativestress prokaryotes have been shown to induceabout 40 proteins (Demple, 1991). Several stronglyaccumulated polypeptides like phytochelatins andmetallothioneins and heat shock proteins at Cd, Cu,As treatment have been reported (Grill et al., 1985;Zenk, 1996; Ma et al., 2003). The decreased proteincontents with the incubation time of CdCl2 ob-

served in our present study (Fig. 7) could ascribe tothe stimulation of protease activity (Siedlecka andKrupa, 2002) and the ending of biosynthesis of somekinds of proteins. The synthesis of phytochelatinsfrom glutathione by phytochelatin synthase, aspecific-glutamylcysteine dipeptidyl transpepti-dase, started within a few minutes after Cd supply,but the reaction ended unless further Cd wassupplied (Loeffler et al., 1989). Otherwise, thispattern of protein in response to Cd stress could berelated to garlic seedlings developing from clovesnot seeds. For example, the protein content inleaves increased about 20% within 6 h of exposureto Cd, such increase may partly be due to thetransportation of nitrogen or soluble proteins fromcloves to shoot. It was reported that Allium cepaplants developing from bulbs were more tolerant toPb than those developing from seeds (Michalak andWierzbicla, 1998).

Taken together, our results suggest that theaccumulation of Cd in roots and leaves promotesthe production of AOS, demonstrated by lipidperoxidation, variation of activity of antioxidativeenzymes, and that these enzymes can cooperatebetween each other, especially between POD andCAT, in the detoxification of AOS. These temporalobservations imply that heavy metal resistance ischaracteristic of garlic seedlings and it is possiblethat there exists at least one evocator that triggersa series of reactions in leaves before Cd accumula-tion.

Acknowledgments

This work was supported by the National ScienceFoundation Council of PRC (Grant no. 30170086,30370127), the Special Project of TransgenicResearch (JY03A2001) and Hi-tech Research andDevelopment Program of China (2001AA645010-5).

References

Alvarez ME, Lamb C. Oxidative burst mediated defenseresponse in plant disease resistance. In: ScandaliosJG, editor. Oxidative stress and the molecular biologyof antioxidant stress and the molecular biology ofantioxidant defenses. New York: Cold Spring HarborLaborotory; 1997. p. 815–39.

Barber JM. Catalase and peroxidase in primary leavesduring development and senescence. Z Pflanzenphy-siol 1980;97:135–44.

Beauchamp C, Fridovich I. Superoxide dismutase: im-proved assay and an assay applicable to acrylamidegels. Anal Biochem 1971;44:276–87.

ARTICLE IN PRESS

H. Zhang et al.984

Becher M, Hofner W. Demonstration of a Cd-complexingcompound in Scenedesmus subspicatus and Zea maysL. similar to phytochelatins. Z Pflanzen Bodenkunde1994;157:87–92.

Bradford MM. A rapid and sensitive method for thequantitation of microgram quantities of proteinutilizing the principle of protein-dye binding. AnalBiochem 1976;72:248–54.

Cuypers A, Vangronsveld J, Clijsters H. Peroxidases inroots and primary leaves of phaseolus vulgaris Copperand Zinc Phototoxicity: a comparison. J Plant Physiol2002;189:869–76.

Degraeve N. Carcinogenic, teratogenic and mutageniceffects of cadmium. Mutat Res 1981;86:115–35.

Demple B. Regulation of bacterial oxidative stress genes.Annu Rev Genet 1991;25:315–37.

Di Cagno R, Guidi L, Stefani A, Soldatini GF. Effects ofcadmium on growth of Helianthus annus seedlings:physiological aspects. New Phytol 1999;144:65–71.

Drazkiewicz M, Tukendorf A, Baszynski T. Age-dependentresponse of maize leaf segments to cadmium treat-ment: effect on chlorophyll fluorescence and phyto-chelatin accumulation. J Plant Physiol2003;160:247–54.

Duxbury T. Ecological aspects of heavy metal responses inmicro-organisms. Adv Microb Ecol 1985;8:185–235.

Grill E, Winnacker EL, Zenk MH. Phytochelatins: theprincipal heavy metal complexing peptides of higherplants. Science 1985;230:674–6.

Heath RL, Packer L. Photoperoxidation in isolatedchloroplasts, I: kinetics and stoichiometry of fattyacid peroxidation. Arch Biochem Biophys1968;125:189–98.

Hendry GAF, Brocklebank J. Iron induced oxygen radicalmetabolism in waterlogged plants. New Phytol1985;101:199–206.

Johnson MD, Kenney N, Stoica A, Hilakivi-Clarke L, SinghB, Chepko GR, Clarke R, Sholler PF, Lirio AA, Foss C,Reiter R, Trock B, Paik S, Martin MB. Cadmium mimicsthe in vivo effects of estrogen in the uterus andmammary gland. Nat Med 2003;9:1081–4.

Kahle H. Response of roots of trees to heavy metals.Environ Exp Bot 1993;33:99–119.

Karpinski S, Reynolds H, Karpinska B, Wingsle G, CreissenG, Mullineaux P. Systemic signaling and acclimation inresponse to excess excitation energy in Arabidopsis.Science 1999;284:654–7.

Lidon FC, Teixeira MG. Oxy radicals production andcontrol in the chloroplast of Mn-treated rice. PlantSci 2000;152:7–15.

Lin CC, Kao CH. Effect of NaCl stress on H2O2 metabolismin rice leaves. Plant Growth Regul 2000;30:151–5.

Loeffler S, Hochberger A, Grill E, Gekeler W, WinnackerEL, Zenk MH. Termination of the phytochelatinsynthase reaction through sequestration of heavymetals by the reaction product. FEBS Lett1989;258:42–6.

Ma M, Lau PS, Jia YT, Tsang WK, Lam SKS, Tam NFY, WongYS. The isolation and characterization of Type 1

metallothionein (MT) cDNA from a heavy-metal-tolerant plant, Festuca rubra cv. Merlin. Plant Sci2003;164:51–60.

Michalak E, Wierzbicla M. Differences in lead tolerancebetween Allium cepa plants developing from seedsand bulbs. Plant Soil 1998;199:251–60.

Mittal R, Dubey RS. Behaviour of peroxidases inrice: changes in enzyme activity and isoforms inrelation to salt stress. Plant Physiol Biochem 1991;29:31–40.

Piqueras A, Olmos E, Martinez-Solano JR, Hellin E. Cd-induced oxidative burst in tobacco BY2 cells: timecourse, subcellular location and antioxidant response.Free Radic Res 1999;31:33–8.

Schickler H, Caspi H. Response of antioxidant enzymes tonickel and cadmium stress in hyperaccumulator plantsof the genus Alyssum. Physiol Plant 1999;105:39–44.

Schutzendubel A, Schwanz P, Teichmann T, Gross K,Langenfeld-Heyser R, Godbold D, Polle A. Cadmium-induced changes in antioxidative systems, H2O2 con-tent and differentiation in pine (Pinus sylvestris)roots. Plant Physiol 2001;127:887–98.

Schutzendubel A, Nikolova P, Rudolf C, Polle A. Cadmiumand H2O2-induced oxidative stress in Populus �

canescens roots. Plant Physiol Biochem2002;40:577–84.

Shaw BP. Effects of mercury and cadmium on theactivities of antioxidative enzymes in the seedling ofPhaseolus aureus. Bio Plant 1995;37:587–96.

Siedlecka A, Krupa Z. Functions of enzymes in heavymetal treated plants. In: Prasad MNV, Kazimierz S,editors. Physiology and biochemistry of metal toxicityand tolerance in plants. Netherlands: Kluwer; 2002. p.314–7.

Stroinski A, Zielezinska M. Cadmium effect on hydrogenperoxide, glutathione and phytochelatin levels inpotato tuber. Acta Physiol Plant 1997;19:127–35.

Tripathi RD, Smith S. Effect of chromium (VI) on growth,pigment content, photosynthesis, nitrate reductaseactivity, metabolic nitrate pool and protein content induckweed Spirodela polyrrhiza L. In: Yunus M, editor.ICPEP, 96 Book of Abstracts, 1996. p. 159.

Vajpayee P, Tripathi RD, Rai UN, Ali MB, Singh SN.Chromium (VI) accumulation reduces chlorophyllbiosynthesis, nitrate reductase activity and proteincontent in Nymphaea alba L. Chemosphere2000;41:1075–82.

Vitoria AP, Lea PJ, Azevedo RA. Antioxidant enzymesresponses to cadmium in radish tissues. Phytochem-istry 2001;57:701–10.

Zenk MH. Heavy metal detoxification in higher plants—areview. Gene 1996;179:21–30.

Zhang J, Cui S, Li J, Kirkham MB. Protoplasmic factors,antioxidant responses, and chilling resistance inmaize. Plant Physiol Biochem 1995;33:567–75.

Zhu YL, Pilon-Smits AH, Jouanin L, Terry N. Overexpres-sion of glutathione synthetase in Indian Mustardenhances cadmium accumulation and tolerance. PlantPhysiol 1999;119:73–9.