determination of oxidized and reduced glutathione, by capillary zone electrophoresis, in brassica...

Jorge MendozaPaula SotoInés AhumadaTatiana Garrido

Departamento de QuímicaInorgánica y Analítica,Facultad de CienciasQuímicas y Farmacéuticas,Universidad de Chile,Santiago, Chile

Determination of oxidized and reduced glutathione,by capillary zone electrophoresis, in Brassicajuncea plants treated with copper and cadmium

A rapid method of capillary zone electrophoresis is described to determine the oxi-dized (GSSG) and reduced (GSH) form of glutathione in plant tissue. In order to sepa-rate both analytes in a fused-silica capillary, the pH and composition of the electrolytesolution were optimized. The electrolyte composition was 100 mmol/L, borate25 mmol/L Tris, and 0.2% w/v metaphosphoric acid (MPA), pH 8.2. Some instrumentalconditions used to run the samples were hydrostatic injection for 30 s, 30 kV appliedvoltage, and UV detection (185 nm) at 257C. Linearity and useful range obtained for thecalibration curves were optimum, with correlation coefficients about 0.999 in the0–120 mmol/L range. The migration time was highly reproducible, less than 5 min beingafforded to run a sample. Electrolyte buffer and samples required a careful pH controlfor optimal separation of both analytes. This aspect constitutes a critical analytical stepwhen acids are used in the procedure for sample preparation. Simultaneous analysis ofGSH and GSSG may provide a useful tool for comparative studies of plants in order toselect those species with a potential capacity for detoxification from toxic elements orthose appearing promising from phytoremediation for these elements.

Keywords: Capillary zone electrophoresis / Heavy metals / Oxidized glutathione / Phytotoxicity /Reduced glutathione DOI 10.1002/elps.200305759

1 Introduction

Heavy-metal phytotoxicity may trigger a variety of adap-tative responses in plants detoxification by chelationbeing a fairly common mechanism [1]. Glutathione per-forms several roles in plants, such as storage and trans-port of reduced sulfur, as a precursor in protein andnucleic acid synthesis, and as a modulator of the activityof several enzymes [2]. In addition, glutathione isacknowledged as one of the precursors in the synthesisof phytochelatins, which are polypeptides rich in thiolgroups, capable of binding heavy metals for theirsequestration. Qualitative and quantitative analysis ofglutathione may provide useful information about plantresponse to high concentrations of heavy metals in theenvironment [3]. Detailed and updated reviews of differ-ent methods for quantitative analysis of reduced (GSH)and oxidized (GSSG) forms of glutathione have recentlybeen published [4, 5]. The most commonly used meth-

ods involve HPLC associated to photometric and fluori-metric detection; however, in many cases, several pre-treatments are required before injection of the sampleinto the column. In this respect, electrochemical detec-tion makes these previous steps unnecessary and offershigher specificity to the analysis of redox reactive com-pounds.

In the last years, some methods have been developedusing capillary electrophoresis (CE) technique whichallow to determine simultaneously both forms of glu-tathione [6–11]. In this respect, CE has some advanta-ges compared with traditional methods, namely goodreproducibility, simplicity of procedure, short time ofanalysis, low injection volume, and low cost of analyses[12]. Some authors have described mainly the deter-mination of reduced species [13–15] or that of both ana-lytes, GSH and GSSG, in animal fluids or tissues [6,8–10]. Studies on plant samples are, however, scarce[11, 16].

Borate buffer is the most widely used background elec-trolyte (BGE), while phosphate [16, 17], acetate [6] andbicarbonate [14] are used to a lesser extent. In somecases, these electrolytes have been modified in orderto optimize separation, migration time [18] or to stackthe sample directly in the capillary [19, 20]. BGEs withphotometric detector have been mostly used, however,

Correspondence: Dr. Jorge Mendoza, Facultad de CienciasQuímicas y Farmacéuticas, Universidad de Chile, Olivos 1007,Casilla 233, Santiago, ChileE-mail: [email protected]: 156-2-7370567

Abbreviations: AAS, atomic absorption spectroscopy; GSH,reduced glutathione; GSSG, oxidized glutathione; MPA, meta-phosphoric acid

890 Electrophoresis 2004, 25, 890–896

2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Electrophoresis 2004, 25, 890–896 CE determination of glutathione in plants 891

to improve sensitivity, other detection methods such aslaser-induced fluorescence [15, 17, 21], mass spectrom-etry [14], or electrochemical detection [13, 22] have beensuccessfully applied. The use of borate as BGE has beenstudied with good results in the 7.8–10.0 pH range andat concentrations ranging from 20 to 300 mmol/L [7, 9,14]. In some cases, low concentrations of Tris bufferhave been incorporated to the electrolyte in order toincrease the power of borate buffer [8, 10]. On the otherhand, different procedures have been described to pre-pare the sample in order not to alter the equilibrium be-tween the oxidized and reduced forms of glutathione,such as 5,5’-dithiobis-2-nitrobenzoic acid [23] or, morerecently, membrane filtration [9]. However, most of theprocedures have made use of acids such as metapho-sphoric (MPA), trichloroacetic, phosphoric, perchloric, orsulfosalicylic acid in order to homogenize the tissuesamples to precipitate protein and minimize oxidativechanges [9–11]. Nevertheless, little has been discussedon modifications of the sample matrix resulting fromthese acids or on BGE ability to buffer the sample atlow pH.

Copper (Cu) is an essential element playing several rolesin plants. However, several changes have been observedin the plant due to phytotoxic levels of the metal, amongthese, the formation of different structures between Cu(II)and cysteine residues of phytochelatins and metallothio-neins [24], decrease in the content of GSH related tochanges in the activity of enzymes in the glutathione cycle[25], and changes in the GSSG/GSH ratio in Phaseolusvulgaris L. [26]. On the other hand, cadmium (Cd) is anelement nonessential to the plant metabolic processes,even though it is readily absorbed both by the root andby the leaves. Cd may induce the formation of metal-lothioneins and phytochelatins, strongly interacting withsulfydryl and phosphate groups [27]. Increased levels ofCd have been associated to an increased protecting roleof GSH; particularly, an increased tolerance to Cd in Bras-sica juncea has been related to an increased glutathionecontent in the plant [28, 29].

B. juncea is characterized by rapid and abundant bio-mass production and some varieties or ecotypes aredescribed as heavy metal-accumulating plants, espe-cially in relation to Cd [30, 31]. Other varieties are consid-ered noncumulative but they can absorb one metal oranother at concentrations higher than physiological with-out exhibiting phytotoxicity symptoms [32]. The purposeof this study was to optimize a method by capillary zoneelectrophoresis in order to identify and quantify the oxi-dized and reduced form of glutathione in foliar tissue. Inaddition, the GSH and GSSH distribution in B. juncealeaves treated with Cu and Cd was studied.

2 Materials and methods

2.1 Chemicals

MPA, boric acid, Tris, and sodium hydroxide used in CE,Cu and Cd standards and all the other analytical-gradechemicals were obtained from Merck (Darmstadt, Ger-many). GSH and GSSG (analytical grade), were obtainedfrom Sigma Chemicals (St. Louis, MO, USA) and wereutilized as received. For all the experiments, Milli-Q qual-ity grade water (Millipore, Bedford, MA, USA) was uti-lized.

2.2 Instrumentation and analytical conditions

All analyses were carried out in a Waters Capillary IonicAnalyzer (Waters Milford, MA, USA) equipped with aphotometric detector. Direct spectrophotometric detec-tion was used, with the detector set at 185 nm. A60 cm675 mm ID AccuSep fused-silica capillary (at257C) was employed. Sample injection was hydrostati-cally carried out, with a 10 cm inlet-outlet capillary heightdifference for 30 s. Various running voltages were triedfrom 10 to 30 kV. Data were collected and analyzedusing the Millenium data analysis software (Waters). TheBGE was daily prepared by mixing 20 mL boric acid(500 mmol/L), 5 mL Tris (500 mmol/L), and 10 mL MPA(2% w/v). Different electrolyte pH values were tried,from 7.8 to 8.6. In all cases, the pH was measured in aPMX3000 pH meter (WTW, Germany) with a combinedelectrode. The required pH was adjusted with NaOH toa final volume of 100 mL. Before use, the solution wasfiltered through a nitrate cellulose membrane (0.45 mm)and degassed with nitrogen for 2 min. The solution elec-tric conductivity was measured with a model LF539 con-ductivity meter (WTW). In order to ensure migration timereproducibility, the capillary was rinsed between runs byconsecutive use of 0.1 mol/L NaOH, water, and BGE for1 min. Standard solutions of 1000 mmol/L GSH andGSSG and the mixture of both analytes were preparedfrom the pure reagents, in 2% MPA weekly and usedimmediately or frozen at 2257C in polyethylene vials untiluse. Standard solutions older than a week were dis-carded.

2.3 Plant growth and metal determination

Brassica juncea seedlings were grown from seeds germi-nated in river sand. After twelve days of growth, seedlingswith uniform size and appearance were chosen, andtransferred to plastic pots containing 1 L nutritive solutionwith the following composition: N 7.2 (8:1 NO3

2 : NH41), K

1.5, Ca 2.2, Cl 0.6, S 0.2, Mg 0.5, and Na 0.1 mmol/L; Fe(as Fe-EDTA [ferric hydroxyethylethylenediaminetriace-


CE

and

CE

C

892 J. Mendoza et al. Electrophoresis 2004, 25, 890–896

tate]) 19.5, B 15.0, Mn 5.4, P 4.0, Zn 1.4, Mo 0.5, and Cu0.4 mmol/L [33]. In order to study the effect of Cu and Cd,the nutritive solution was enriched with Cu(NO3)2 orCd(NO3)2 to obtain a 2 mmol/L concentration of each ele-ment above the control concentration. All solutions wereadjusted at pH 6.0 with NaOH. The pots were kept for 28days in a growth chamber under controlled light, temper-ature, and moisture.

2.4 Sample collection and preparation

After the growth period the plants were removed from thesolutions, washed thoroughly with distilled water, oven-dried at 607C for 48 h, was mill-ground, and digestedwith HNO3:H2O2 2:1 in an mls1200 model microwaveoven (Milestone). Then the elements were determined byatomic absorption spectroscopy (AAS) in a 1100B modelatomic absorption spectrophotometer (Perkin Elmer, Nor-walk, CT, USA). One day before harvesting the plants, 1 gof fresh shoot was collected from each pot in order todetermine GSH and GSSG foliar content. The sampleswere stored in cryogenic vials under liquid N2. The frozensamples were crushed in a mortar and then 2% MPA wasslowly added with stirring. The resulting extract was cen-trifuged at 12 000 rpm for 5 min in a cooled centrifuge at57C and the supernatant filtered through a 0.45 mm cellu-lose nitrate membrane. The resulting solutions wereimmediately analyzed.

2.5 Statistical analysis

Results were analyzed by one-way analysis of variance(ANOVA) and the mean values of the various treatmentswere compared by Duncan’s multiple range test to a levelof 5%.

3 Results and discussion

Considering that several GSH and GSSH analytical meth-ods by CE use acids for extraction, our method assessedthe effect of MPA incorporation to the BGE in order todiminish interference of the acid on the sample electro-phoretic signals and to improve linearity and reproducibil-ity of results. Taking existing information [10] into account,we worked with a mixture containing borate and Tris as aBGE, modified by adding MPA and varying its pH. A posi-tive power source was used, the polarity of which deter-mined that the cathode was located at the outlet end ofthe capillary with the direction of the electrosmotic flow(EOF) towards the negatively charged cathode. Underthese conditions, cations will migrate first, followed byneutral molecules and water and finally by both forms of

glutathione, which will be negatively charged at pH 8.2and will be carried by the EOF of higher magnitude andopposite sense to the attractive force between suchmolecules and the anode [12].

At a previous stage, an electrolyte with a borate concen-tration increased from 50 to 200 mmol/L was tested,keeping the Tris concentration unchanged (25 mmol/L).The best conditions were obtained for 100 mmol/L with acurrent of 76 mA and a migration time of about 6 min.Incorporation of 0.2% MPA increased the ionic strengthof the electrolyte, producing a conductivity of 3.6 mS/cmand a current of about 130 mA. With 0.3% MPA, conduc-tivity and current increased and 4.5 mS/cm and 170 mAwere obtained, respectively, resulting in a marked de-crease of the baseline noise. Higher concentrations gen-erated currents higher than 200 mA, with a risk of generat-ing microbubbles which would cause current cutoffs inthe system [12]. For this reason, the chosen electrolytewas a mixture of 100 mmol/L borate, 25 mol/L Tris, and0.2% MPA.

The electrolyte pH is a key factor in CE signal resolution. Ahigher pH implies a higher Z potential (z) in the capillarywall and consequently a higher electrosmotic rate. Onthe other hand, a higher pH means a higher degree of ioni-zation of solutes such as glutathione and hence a higherelectrophoretic rate [11, 12]. In our experiment, signalseparation and performance was evaluated by varyingthe BGE pH within the 7.8–8.6 range (Fig. 1). At pH 7.8and 8.0 only the GSSG signal was observed. In the 8.1–8.6 pH range both signals were observed, the sharpestones at 8.2 and 8.6 but in the latter case the generatedcurrent was too high (202–204 mA) with the risk of gener-ating microbubbles in the capillary.

In those studies where the use of acids is described inorder to extract both glutathione forms, there is littleinformation on the effect of sample pH on signal separa-tion. Only Carlucci et al. [7], who used phosphoric acid,briefly mentioned that sample neutralization is a criticalanalytical step. In our case, the pH of standard solutioncontaining GSH and GSSG in MPA was 1.9. On runningthis solution with the above-mentioned electrolyte, comi-gration of analytes was observed (Fig. 2). However, uponaddition of NaOH (1 mol/L) in order to neutralize the so-lution, the resolution of signals improved gradually withincreasing alkali volume. Optimum pH for the separationof the standards prepared in the MPA matrix was 8.3,and above this value analyte comigration was againobserved.

The changes observed in migration times and signalseparation with increasing electrolyte pH may beaccounted for by changes in ionization of the terminal glu-



Figure 1. Effect of BGE pH on migration time and perfor-mance of signals for GSH and GSSG running under opti-mized conditions.

Figure 2. Effect of sample pH on separation and migra-tion time of GSH and GSSG signals.

tamate amino functional group (pKa , 8.2) as describedby Davey et al. [11]. An additional explanation would behigher borate and Tris ionization, which produces anincrease in the electrolyte dielectric constant and conse-quently of z [12]. Considering this effect, our procedureimplied neutralizing 200 mL of sample or standard by



adding 60 mL of 1 mol/L NaOH to reach a pH of 8.3 6

0.1. This procedure was carried out by slowly addingthe alkali to the same measuring vial while this was keptunder vortex swirling. The resulting solution was immedi-ately run. Under these conditions, migration times of4–5 min were observed for both analytes in standardsand samples (Fig. 3). Detection limits for both analyteswere considered as the analyte concentration providinga signal equal to the blank signal plus three times theblank standard deviation, daily determined for three con-secutive days. The resulting values were 6.1 mmol/L(CV% = 2.0) for GSH and 5.2 mmol/L (CV% = 3.3)for GSSG. The relative electrophoretic mobilities forGSH and GSSG were 22.3961028 m2V21s21 and22.4661028 m2V21s21, respectively. At the same time,assessment was done of the linear correlation betweenconcentration of each analyte and corresponding peakarea. Using a concentration of up to 120 mmol/L of eachanalyte provided remarkable linearity, with a correlationcoefficient of 0.999 or higher every day. The best calibra-tion curves were Y = 121.09X – 526.17 (r = 0.9994) forGSH and Y = 304.05X – 332.93 (r = 0.9995) for GSSG.Additionally, increasing the range of the calibration curveto 150 mM showed good linearity with coefficients above0.99. Migration times were 4.407 min (n = 5, CV% = 0.3)and 4.756 min (n = 5 CV% = 0.2) for GSH and GSSG,respectively.

In order to assess the separation efficiency of the method,the recovery of both analytes was tested in the samplematrix. The samples were spiked with 50 mmol/L GSH

Figure 3. Electropherogram of a sample of control plantsobtained for extraction of leaves of Brassica juncea with2% MPA.

and GSSG at a stage prior to centrifugation and the ana-lytes were determined by the established method, with97% (n = 5) and 101% (n = 5) recovery for GSH andGSSG, respectively. In addition, and in order to verifyGSSG and GSH signals, some of the samples and stand-ards were run with addition of mercaptoethanol so as toreduce GSSG to GSH (Fig. 4).

Considering the results of Carru et al. [9] on oxidative sta-bility of both forms of glutathione, some samples werekept at 47C and analyzed 24 and 48 h after extraction.Initially, observed GSH and GSSG levels did not changewith time, a fact that, associated to the good recoveryobtained by this method, would delete any effect of 2%MPA on the oxidation of these analytes in a two-day peri-od. In our experiment with plants, dry matter yield andshoot appearance were not affected by Cu (Table 1).However, plants treated with Cd showed phytotoxicitysymptoms expressed as low shoot development andslight chlorosis. Cd caused a significant decrease in leafdry matter yield compared with Cu-treated plants andcontrols. Both metals showed no significant effect onroot growth. Plants treated with Cu showed a metal con-

Figure 4. Effect of mercaptoethanol (ME) on GSH andGSSG signals. (A) Standard containing GSH and GSSG(100 mmol/L) without ME. (B) Same standard with additionof 5 mL of ME.



Table 1. Dry matter yield and concentration of Cu, Cd, GSH, and GSSG obtained in the experiment(n = 4)

Treatment

Control Copper Cadmium

Dry matter yield (g): Shoot 1.762 (60.43)a) ab) 1.795 (60.61) a 0.765 (60.05) bRoot 0.198 (60.05) b 0.217 (60.09) a 0.112 (60.01) c

Copper (mmol/g): Shoot 0.11 (60.01) c 0.37 (60.06) a 0.22 (60.02) bRoot 0.44 (60.06) c 3.14 (60.55) a 1.52 (60.19) b

Cadmium (mmol/g): Shoot ndc) nd 1.87 (60.09)Root nd nd 11.48 (60.92)

GSH (mmol/g) Shoot 0.75 (60.05) a nd 0.41 (60.05) b

GSSG (mmol/g) Shoot 1.24 (60.08) a 0.82 (60.11) b 0.67 (60.04) b

a) Standard deviation in parenthesesb) Means within a line followed by the same letter do not differ significantly (p , 0.05).c) nd, not detected

centration 2 and 6 times higher than that observed in thecontrol plants for shoot and root, respectively. Similarly, asharper increase was observed for Cd which, consider-ing the AAS detection limit, was estimated at a concen-tration 110 and 410 times higher, respectively. In addi-tion, Cd-treated plants showed a Cu concentration inshoot and root with values fluctuating from those in thecontrol plants to those in Cu-treated plants, indicatinga possible alteration in the process of Cu absorption asan effect of Cd. GSH and GSSG concentrations ob-served in the leaves of B. juncea were significantly lowerin Cu- and Cd-treated plants compared with controlplants (Table 1). This decrease could be due to the factthat both peptides are utilized as precursors in the syn-thesis of cysteine-rich peptides such as metallothioneinsand phytochelatins, constituting two important mechan-isms of metal detoxification. Another possible explana-tion could be an effect of Cu and Cd on enzymatic activ-ities related to the glutathione cycle, by altering GSH andGSSG levels. In this respect, Zhu et al. [34] suggestedthat g-glutamil-cysteinesynthetase and glutathione syn-thetase enzymes inhibit GHS production under Cdstress. In relation to Cu, Xiang and Oliver [35] have foundan increase in oxidative stress due to Cu12 treatment,expressed as an increase of 2–5 times in GSSG com-pared with GSH in Arabidiopsis sp. Similarly, an altera-tion has been described in the levels of both peptidesusing the GSH/(GSH10.5GSSG) ratio as an indicator ofoxidative stress in algae, and a decrease of 0.5–0.2 hasbeen reported in this ratio by increasing the level of Cu[36]. Applying this criterium, the ratio of the experimentshere described decreased from 1.5 to 0.8 as an effect ofCu, the same as described by Rijstenbil et al. [36]. In thissense, no changes were observed in the ratio as aneffect of Cd.

4 Concluding remarks

The study of interactions between metals and biomole-cules in plants constitutes an important challenge inwhich CE is relevant on account of its versatility in simul-taneous analysis of metal forms or molecules coexistingin the same matrix. The method here described permitsdirect measurement of GSH and GSSG in plant tissuewithin a short time and with good reproducibility. Includ-ing the time spent in sample preparation, the whole anal-ysis takes about 20 min per sample, which makes thismethod applicable to studies with numerous samples orin routine analyses. Linearity and useful range of the cali-bration curve were excellent, permitting the accurateanalysis of samples with concentrations of both peptidesof up to 150 mmol/L in the extract without further dilution.Migration times were highly reproducible, less than 5 minbeing required to run a sample. On the other hand, sam-ple and BGE pH control was a critical analytical step, con-stituting an insufficiently described aspect in other stud-ies using acids for extraction. As to acid-extracted sam-ple stability, our results indicate that the use of MPA inglutathione analysis does not produce oxidative changesin the GSH and GSSG levels, unlike the results describedby Carru et al. [9]. Finally, the simultaneous detection ofboth peptides of glutathione may constitute a useful toolfor comparative studies of plants grown under controlledconditions in order to select those species with a potentialcapacity to detoxify from toxic elements or those appear-ing promising for phytoremediation for these elements.

This work was supported by the Facultad de Ciencias Quí-micas y Farmacéuticas, Universidad de Chile (ProjectMemorias 2000).

Received July 28, 2003



5 References

[1] Cobbett, C., Curr. Opin. Plant Biol. 2000, 3, 211–216.

[2] May, M., Vernoux, T., Leaver, C., Van Montagu, M., Inzé, D.,J. Exp. Bot. 1998, 49, 649–667.

[3] Sutter, K., Friebe, S., Krauss, G. J., Int. J. Environ. Anal.Chem. 1999, 74, 167–178.

[4] Lock, J., Davis, J., Trends Anal. Chem. 2002, 21, 807–815.

[5] Camera, E., Picardo, M., J. Chromatogr. B 2002, 781, 181–206.

[6] Kratumacher, C., Schilling, D., Pittelkow, M. R., Naylor, S.,Biomed. Chromatogr. 2002, 16, 224–228.

[7] Carlucci, F., Tabucchi, A., Biagioli, B., Sani, G., Lisi, G., Mac-cherini, M., Rosi, F., Marinello, E., Electrophoresis 2000, 21,1552–1557.

[8] Serru, V., Baudin, B., Ziegler, F., David, J. P., Cals, M. J., Vau-bourdolle, M., Mario, N., Clin. Chem. 2001, 47, 1321–1324.

[9] Carru, C., Zinellu, A., Pes, G. M., Marongiu, G., Tadolini, B.,Deiana, L., Electrophoresis 2002, 23, 1716–1721.

[10] Muscari, C., Pappagallo, M., Ferrari, D., Giordano, E.,Capanni, C., Caldarera, C. M., Guarnieri, C., J. Chromatogr.B 1998, 707, 301–307.

[11] Davey, M. W., Bauw, G., Van Montagu, M., J. Chromatogr. B1997, 697, 269–276.

[12] Baker, D. R., Capillary Electrophoresis, John Wiley & Sons,New York 1995, pp. 1–52.

[13] Wang, W., Xin, H., Shao, H. L., Jin, W. R., J. Chromatogr. B2003, 789, 425–429.

[14] He, T., Quinn, D., Fu, E., Wang, Y. K., J. Chromatogr. B 1999,727, 43–52.

[15] Lochman, P., Adam, T., Friedecky, D., Hlídková, E., Skop-ková, Z., Electrophoresis 2003, 24, 1200–1207.

[16] Herrero-Martinez, J. M., Simó-Alfonso, E. F., Ramis-Ramos,G., Deltoro, V. I., Calatayud, A., Barreno, E., Environ. Sci.Technol. 2000, 34, 1331–1336.

[17] Toyo’oka, T., Tanabe, J., Kashihara, Y., Anal. Chim. Acta2001, 433, 1–12.

[18] Vegvari, A., Larsson, A. K., Hjertén, S., Mannervik, B.,ChemBioChem 2002, 3, 1117–1125.

[19] Shihabi, Z. K., Hinsdale, M. E., Cheng, C. P., Electrophoresis2001, 22, 2351–2354.

[20] Kong, Y., Zheng, N., Zhang, Z. C., Gao, R. Y., J. Chromatogr.B 2003, 795, 9–15.

[21] Zinellu, A., Carru, C., Galistu, F., Usai, M. F., Pes, G. M., Bag-gio, G., Federici, G., Deiana, L., Electrophoresis 2003, 24,2796–2804.

[22] Jin, W. R., Li, X. J., Gao, N., Anal. Chem. 2003, 75, 3859–3864.

[23] Mills, B. J., Richie, J. P., Lang, C. A., Anal. Biochem. 1990,184, 263–267.

[24] Kabata Pendias, A., Trace Elements in Soil and Plants, CRCPress, Boca Raton, FL 2000, pp. 99–122.

[25] Nagalakshmi, N., Prasad, M. N. V., Plant Sci. 2001, 160,291–299.

[26] Cuypers, A., Vangronsveld, J., Clijsters, H., Physiol. Plant2000, 110, 512–517.

[27] Meagher, R., Plant Biol. 2000, 3, 153–162.

[28] Zaroogian, G., Jackim, E., Comp. Biochem. Physiol. C-Phar-macol. Toxicol. Endocrinol. 2000, 127, 251–261.

[29] Pilon-Smits, E., Zhu, Y. L., Sears, T., Terry, N., Physiol. Plant2000, 110, 455–460.

[30] Zhu, Y. L, Pilon-Smits, E., Jouanin, L., Terry, N., Plant Phy-siol. 1999, 119, 73–79.

[31] Salt, D. E., Prince, R. C., Pickering, I. J., Raskin, I., PlantPhysiol. 1995, 109, 1427–1433.

[32] Salt, D. E., Smith, R. D., Raskin, I., Plant Physiol. Plant Mol.Biol. 1998, 49, 643–648.

[33] Mendoza, J., Borie, F., Commun. Soil Sci. Plant Anal. 1998,29, 681–695.

[34] Zhu, Y. L., Pilon-Smits, E., Tarum, A., Weber, S., Jouanin, L.,Terry, N., Plant Physiol. 1999, 121, 1169–1177.

[35] Xiang, C., Oliver, D. J., Plant Cell 1998, 10, 1539–1550.

[36] Rijstenbil, J. W., Haritonidis, S., Malea, P., Seferlis, M., Wijn-holds, J. A., Hydrobiologia 1998, 385, 171–181.


determination of oxidized and reduced glutathione, by capillary zone electrophoresis, in brassica...

Documents