Cadmium effect on oxidative metabolism of pea (Pisum sativum L.) roots. Imaging of reactive oxygen species and nitric oxide accumulation in vivo


  • Sequence data from this article has been deposited with the EMBL/GenBank data libraries under accession number AY426764.

Luisa M. Sandalio. Fax: +34 958129600; e-mail:


Growth of pea (Pisum sativum L.) plants with 50 µm CdCl2 for 15 d produced a reduction in the number and length of lateral roots, and changes in structure of the principal roots affecting the xylem vessels. Cadmium induced a reduction in glutathione (GSH) and ascorbate (ASC) contents, and catalase (CAT), GSH reductase (GR) and guaiacol peroxidase (GPX) activities. CuZn-superoxide dismutase (SOD) activity was also diminished by the Cd treatment, although Mn-SOD was slightly increased. CAT and CuZn-SOD were down-regulated at transcriptional level, while Mn-SOD, Fe-SOD and GR were up-regulated. Analysis of reactive oxygen species (ROS) and nitric oxide (NO) levels by fluorescence and confocal laser microscopy (CLM) showed an over-accumulation of O2.− and H2O2, and a reduction in the NO content in lateral and principal roots. ROS overproduction was dependent on changes in intracellular Ca+2 content, and peroxidases and NADPH oxidases were involved. Cadmium also produced an increase in salicylic acid (SA), jasmonic acid (JA) and ethylene (ET) contents. The rise of ET and ROS, and the NO decrease are in accordance with senescence processes induced by Cd, and the increase of JA and SA could regulate the cellular response to cope with damages imposed by cadmium.


Cadmium is a toxic metal that normally occurs in low concentration in soils but can enter the environment mainly from industrial processes and phosphate fertilizers, and then is transferred to the food chain (Wagner 1993). Cadmium is taken up rapidly by the roots and can be loaded into the xylem for its transport into leaves. The amount of Cd accumulated in roots or translocated to leaves differs considerably among species. Most plants are sensitive to low Cd concentrations, which inhibit root and shoot growth, as a consequence of alterations in the photosynthesis rate, uptake and distribution of macronutrients and micronutrients (Lozano-Rodríguez et al. 1997; Sandalio et al. 2001).

Cadmium can be detoxified by complexation by phytochelatins, a family of GSH-derived peptides that contain a varying number of glutamate and cysteine residues (Zenk 1996). The synthesis of phytochelatins is induced by Cd and other heavy metals, and this is accompanied by a decrease in the tissue concentration of GSH (Zenk 1996). GSH not only is an important antioxidant in the ASC–GSH cycle involved in the removal of hydrogen peroxide in different cell compartments (Jiménez et al. 1997; Noctor et al. 1998), but also has key functions in redox signalling processes (Foyer & Noctor 2003; Foyer, Trebst & Noctor 2006).

Cadmium produces concentration-dependent imbalances in the antioxidant defences of plants and induces oxidative stress (Romero-Puertas et al. 1999, 2002; Dixit, Pandey & Shymar 2001; Sandalio et al. 2001; Schützendübel et al. 2001). In Phaseolus vulgaris, Phaseolus aureus and Helianthus annuus, the toxicity of Cd has been related with the increase of lipid peroxidation and alterations in the antioxidant systems (Somashekaraiah, Padmaja & Prasad 1992; Shaw 1995; Gallego, Benavides & Tomaro 1996; Smeets et al. 2005). Recently, in leaves from pea (Pisum sativum L.) plants grown with CdCl2, the production of ROS at subcellular level was demonstrated (Romero-Puertas et al. 2004). ROS were detected in epidermal, transfer and messophyll cells, the plasma membrane being the main source of ROS, although mitochondria and peroxisomes were also involved (Romero-Puertas et al. 2004). Other possible sources of ROS in lignifying tissues are cell wall peroxidases, as Blee et al. (2003) have demonstrated using tobacco transformant plants deficient in TP60 peroxidase. Although Cd does not participate in Fenton-type ROS-producing reactions, it can indirectly activate NADPH oxidases in membranes, giving rise to oxidative bursts (Piqueras et al. 1999; Romero-Puertas et al. 2004). The enzymes SOD and CAT, and peroxidases are involved in the detoxification of O2.−, and H2O2, respectively, thereby preventing the formation of ·OH radicals. GR as well as GSH are important components of the ASC–GSH cycle responsible for the removal of H2O2 in different cell compartments (Jiménez et al. 1997; Noctor & Foyer 1998).

The gaseous-free radical nitric oxide (NO) is a widespread intracellular and intercellular messenger with a broad spectrum of regulatory functions in many physiological processes. NO was reported to be involved in ethylene (ET) emission (Leshem & Haramaty 1996), response to drought (Leshem 1996), disease resistance (Durner, Wendehenne & Klessig 1998; Clark et al. 2000; Delledonne et al. 2001), growth and cell proliferation (Ribeiro et al. 1999), maturation and senescence (Leshem, Wills & Veng-Va Ku 1998; Corpas et al. 2004), apoptosis/programmed cell death (Magalhaes, Pedroso & Durzan 1999; Pedroso & Durzan 2000; Clark et al. 2000) and stomatal closure (García-Mata & Lamattina 2002; Neill et al. 2002a). There are several enzymes that were shown to produce NO in plants. Nitrate reductase is a well-established generator of NO (Dean & Harper 1988; Rockel et al. 2002), although this enzyme does not produce NO from l-arginine and therefore, it cannot be considered as a characteristic NO synthase (NOS) activity. Another enzyme that can generate NO from nitrite is a plasma membrane-bound enzyme of tobacco roots (Stöhr et al. 2001). In the past years, there has been an increasing number of reports showing the presence of NOS-like activities in plants (for a review, see  del Río, Corpas & Barroso 2004), including an l-arginine-dependent activity localized in peroxisomes that was biochemically characterized (Barroso et al. 1999; Corpas et al. 2004). Recently, an Arabidopsis gene (AtNOS1) that encodes an inducible protein with NOS activity, AtNOS1, was isolated and characterized (Guo, Okamoto & Crawford 2003). This protein has no sequence similarities to known animal NOS enzymes.

In this work, the effect of growing pea plants with CdCl2 on enzymatic and non-enzymatic antioxidants, and different oxidative stress markers, was studied in pea roots. Production of ROS and NO under cadmium stress was investigated in vivo by confocal laser microscopy. In order to get deeper insights into the mechanism of cell response to Cd toxicity, different molecules involved in signalling under metal stress and the expression of enzymatic antioxidants were analysed.


Plant material and growth conditions

Pea (P. sativum L., cv. Lincoln) seeds were obtained from Royal Sluis (Enkhuizen, Holland). Plants were grown in the greenhouse in aerated full-nutrient media under optimum conditions for 14 d (Sandalio et al. 2001). Then, the media either remained unsupplemented (control plants) or were supplemented with 50 µm CdCl2 (Cd-treated plants), and the plants were further grown for 14 d.

Enzymatic assays

Total SOD activity (EC was assayed according to the ferricytochrome c method of McCord & Fridovich (1969). SOD isoenzymes were individualized by native polyacrylamide gel electrophoresis (PAGE) on 10% acrylamide gels and were localized by a photochemical method (Beauchamp & Fridovich 1971). CAT activity (EC was determined as described by Aebi (1984). GR activity was assayed as described by Jiménez et al. (1997) and guaiacol peroxidase (GPX) activity was determined according to Quessada & Macheix (1984).

Semi-quantitative reverse transcriptase PCR (RT-PCR)

A 2.5 µg amount of total RNA from roots was used as a template for the reverse transcriptase (RT) reaction. It was added to a mixture containing 0.2 µm oligo dT23 anchored (Sigma St. Louis, MO, USA), H2O2 diethyl pyrocarbonate until 25 µL, heated at 70 °C for 10 min and placed in ice for 5 min. It was add 1.5 mm dNTPs, 1X RT-buffer, 1 U ribonuclease inhibitor (Sigma), 20 U AMV RT (Finnzymes, Espoo, Finland). The reaction was carried out at 42 °C for 40 min, followed by a 5 min step at 98 °C and then by cooling to 4 °C.

Amplification of actin II cDNA from pea was chosen as a control. The oligonucleotides used in this work are shown in Table 1. cDNAs were amplified by PCR as follows: 1 µL of the produced cDNA diluted (1/10 or 1/20) was added to 0.2 mm dNTPs, 1.5 mm MgCl2, 1X PCR buffer, 1.25 U of Hot Master Taq polymerase (Eppendorf, Hamburg, Germany) and 0.4 µm of each primers (see Table 1) in a final volume of 20 µL. Reactions were carried out in a Master Cycler (Eppendorf). A first step of 2 min at 94 °C was followed by 30 cycles of 20 s at 94 °C, 20–40 s at annealing temperature and 30 s at 65 °C with a final extension of 10 min at 65 °C. Amplified PCR products were detected after electrophoresis in 0.8% agarose gels (Serva, Heidelberg, Germany) stained with ethidium bromide. Quantification of the bands was performed using a Gel Doc system (Bio-Rad, Hercules, CA, USA) coupled with a high sensitive coupling charge device (CCD) camera. Band intensity was expressed as relative absorbance units, and ratio between a specific cDNA and ACTII amplification was calculated to normalize for initial variations in sample concentration.

Table 1.  Oligonucleotides used for the semi-quantitative PCR analysis
GeneOligonuclotide sequence (5′−3′)Product size (bp)Accession No
  1. F, forward oligonucleotide; R, reverse oligonucleotide; GR, glutathione reductase; SOD, superoxide dismutase.

Catalase (CAT)
Actin II

H2O2 determination in roots extracts

H2O2 concentration in crude extracts from pea roots was determined by spectrofluorometry, according to the method described by Romero-Puertas et al. (2004). All operations were performed at 0–4 °C. The roots (0.4 g) were homogenized in 1.2 mL of 25 mm HCl, and the crude extracts were filtered through two nylon layers, and pigments were removed by mixing with 15 mg of charcoal (Sigma). The pigment-containing charcoal was separated by centrifugation at 5000 g for 5 min, and the supernatants were clarified by filtration through a 0.22 µm filter unit. The pH of extracts was adjusted to 7.0 with NaOH and these extracts were used to measure the H2O2 concentration. The reaction mixtures (3 mL) contained 50 mm Hepes buffer, pH 7.6, 5 mm homovanillic acid and 10–100 µL of sample. The reaction was started by adding horseradish peroxidase to a final concentration of 40 µm, and the fluorescence produced was measured in a spectrofluorophotometer (Shimadzu RF-540, Shimadzu, Kyoto, Japan) at excitation and emission wavelengths of 315 and 425 nm, respectively. The H2O2 concentration was determined from a calibration curve of H2O2 (Merck, Rahway, NJ, USA) in the range 0.5–80 µm.

Quantitation of GSH and ASC

For determination of ASC and GSH, the methods of the bipiridyl and oxidation of 5,5′-dithiobis(2-nitrobenzoic acid) (DTNB), respectively, were used (Knörzer, Durner & Böger 1996; Griffith 1980).

ROS and NO detection by fluorescence microscopy

For NO detection, segments of pea roots of approximately 20 mm2 from the apex were incubated for 1 h at 25 °C, in darkness, with 10 µm 4,5-diaminofluorescin diacetate (DAF-2 DA, Calbiochem, San Diego, CA, USA) prepared in 10 mm Tris–HCl (pH 7.4). Superoxide radicals were detected by staining with 10 µm dihydroethidium (DHE) (Yamamoto et al. 2002), and H2O2 was localized using 25 µm 2′,7′-dichlorofluorescin diacetate (DCF-DA) (Tarpey, Wink & Grisham 2004). As negative controls, the pea root segments were incubated with 1 mm aminoguanidine (NOS inhibitor), 1 mm tetramethyl piperidinooxy (TMP) (O2.− scavenger) and 1 mm ASC (H2O2 scavenger). Then, the segments were washed twice in the same buffer for 15 min each and were then embedded in 30% polyacrylamide blocks. Roots sections were mounted and observed in a confocal laser scanning microscope (Leica TCS SL, Leica Microsystems, Wetzlar, Germany). DAF-2 DA and DCF-DA fluoresce green (excitation 495 nm, emission 515 nm and 485 nm excitation, 530 nm emission, respectively) and DHE fluoresces red (490 nm excitation, 520 nm emission).

Detection of cell death

To determine changes in viability of cells by Cd treatment, pea root sections of 20 mm2 were infiltrated with a 0.25% (w/v) aqueous solution of Evan’s Blue (Romero-Puertas et al. 2004), for 5 h.

Treatments with inhibitors

Prior to staining with the fluorescence dyes, pea roots segments from control and Cd-treated plants of approximately 20 mm2 from the apex were incubated for 30 min with the following inhibitors: (1) 5 µm cantharidin (protein phosphatase inhibitor); (2) 1 mm LaCl3 (Ca2+ channel blocker); (3) 10 µm diphenylene iodonium (DPI) (NADPH oxidase inhibitor); (4) 1 mm NaN3 (peroxidase inhibitor); (5) 1 mm TMP (O2.− scavenger); and (6) 1 mm ASC (H2O2 scavenger).

Jasmonic acid (JA) and salicylic acid (SA) determination

JA and SA were quantitated in duplicate as described by Vigliocco et al. (2002). Pea roots (5 g fresh weight) were homogenized in 10 mL of acetone. The extracts were dried by speed vacuum at 40 °C and resuspended in 5% ethyl ether. The ethyl ether fraction was purified by gas chromatography (GC) on a solid-phase-extraction (SPE) C18 column and eluted with 25 mL of ether. The fraction was dried at 25 °C and redissolved in 5 mL of hexane:ether (90:10). This fraction was purified on an SPE SiO2 column eluted with 50 mL of hexane:ether (90:10). This fraction contained free SA and JA. To obtain the SA and JA derived fraction (MeSA and MeJA, respectively), the C18 column was eluted with 50 mL of ether and derivatized with diazomethane. The eluted fractions were quantitated and identified by GC-mass spectrometry (GC-MS) by select ion monitoring (SIM) at m/z 120 and 152 for the MeSA and at m/z 224 and 151 for the MeJA. Standards of MeSA and MeJA were used in the range of 10–500 ng. The GC-MS system was equipped with a 30 × 0.25 × 0.25 m film DB-MS column. Temperatures were as follows: from 40 °C (1 min) to 20 °C min−1 until 150 °C (3 min), and from 5 °C min−1 to 230 °C. Helium was used as carrier gas (1 mL min−1) and splitless injection (1 µL).

ET determination

For determination of endogenous ET production, fresh roots (10–15 g) were placed in 100 mL hermetic vials, flushed with ET-free air and incubated for 2 h at room temperature. The ET concentration was determined on a gas chromatograph (Perkin Elmer 8600, Perkin Elmer, Wellesley, MA, USA) fitted with a flame ionization detector and a Poropak-R column. Nitrogen was used as carrier gas and a commercial standard mixture of ET was used for calibration of the gas chromatograph.

Other assays

Lipid peroxidation was determined by measuring the concentration of thiobarbituric acid-reacting substances (TBARS) as described by Buege & Aust (1972). Carbonyl groups were assayed using the dinitrophenyl hydrazine method, according to Levine et al. (1991). Proteins were determined by the method of Bradford (1976) using bovine serum albumine (BSA) as standard.

Data were subjected to one-way analysis of variance for each parameter. When the effect was significant (P < 0.05), differences among means were evaluated for significance by Duncan’s multiple-range test (P < 0.05).


Effect of Cd2+ on growth, morphology and ultrastructure of roots

In pea plants, cadmium induced a reduction of the root growth, mainly in the number and length of lateral roots (Fig. 1a). A thickening of lateral roots could be also observed in Cd-treated plants, chiefly because of an increase in the number of cortex cells (Fig. 1c), as well as browning of the root tissue. The vascular cylinder also underwent changes mainly in principal roots, thus xylem vessels were distributed in a different way, showing smaller vessels in Cd-treated roots than in control roots (Fig. 1e). Phloem was also apparently less developed in pea roots treated with Cd (Fig. 1c & e).

Figure 1.

Cadmium effect on growth and structure of roots of pea plants. Plants were grown with 50 µm Cl2Cd for 15 d in hydroponic medium. (a) Complete roots from control and Cd-treated pea plants. (b, c) Cross-section of lateral roots from control and Cd-treated pea plants, respectively. (d, e) Cross-section of principal roots from control and Cd-treated pea plants, respectively.
c, cortex; e, epidermis; Pe, pericycle; Ph, phloem; X, xylem.

Parameters of oxidative stress in pea roots

The analysis of enzymatic activity of antioxidants in pea roots showed a significant reduction of GR and GPX and, to a lower extent, of CAT, while total SOD activity showed a slight increase by the metal treatment (Fig. 2a). However, the analysis of the activity of SOD isoforms by native PAGE showed a strong reduction of CuZn-SODs, while Mn-SOD was slightly increased and Fe-SOD was not affected (Fig. 2b). The study by semi-quantitative RT-PCR of the expression of enzymatic antioxidants showed a Cd-induced down-regulation of CAT (40% reduction) and CuZn-SOD (50% reduction), which was parallel to the decrease in enzyme activity described earlier. On the contrary, an up-regulation of Fe-SOD (40% higher), Mn-SOD (70%) and GR (40%) was observed (Fig. 2c), despite the reduction in the activity of GR (Fig. 2a).

Figure 2.

Cadmium effect on antioxidative enzymes in pea roots. (a) Enzymatic activities from crude root extracts. Catalase (CAT) activity was expressed as µmol H2O2 mg−1 protein min−1, glutathione reductase (GR) as nmol NADPH mg−1 protein min−1, and superoxide dismutase (SOD) and guaiacol peroxidase (GPX) as U mg−1 protein. (b) SOD isozyme activities. Proteins (45 µg) were subjected to native polyacrylamide gel electrophoresis (10% polyacrylamide gels) and isozymes were visualized by a photochemical method. (c) Analysis of mRNA expression of CAT and SODs by semi-quantitative reverse transcriptase PCR. Each rectangle represents the mean ± SE of the mean of three replicates. Differences were significant at P < 0.05 (*) according to Duncan’s multiple range test.

The determination of the non-enzymatic antioxidants (ASC and GSH) showed a reduction of the total content of both compounds (oxidized + reduced) by Cd, and the most significant decrease took place in the GSH content (almost a 50% reduction) (Fig. 3). Oxidized ASC (DHA) and reduced GSH were the mains forms of these antioxidants detected in Cd-treated pea roots (Fig. 3a & b).

Figure 3.

Content of (a) ascorbic acid (ASC) and (b) glutathione (GSH) in pea root extracts from control and Cd-treated plants. Each rectangle represents the mean ± SE of the mean of three replicates. Differences were significant at P < 0.05 (*) and P < 0.01 (**) according to Duncan’s multiple range test.
DHA, dehydroascorbic acid; GSSG, oxidized glutathione.

Two markers of oxidative damage to macromolecules, the carbonyl group content of proteins and extent of lipid peroxidation, were measured (Fig. 4). A significant increase in the content of malondyaldehyde (MDA), marker of lipid peroxidation, was observed in roots of plants exposed to Cd, and a slight although not significant increase in protein oxidation was also observed. The determination of H2O2 content in roots extracts showed an increase of two times in plants exposed to Cd (Fig. 4), which could explain the results obtained for the oxidative stress parameters.

Figure 4.

Effect of cadmium treatment of pea plants on lipid peroxidation [measured as malondyaldehyde (MDA) content], protein oxidation (measured as carbonyl group content) and H2O2 concentration. Each rectangle represents the mean ± SE of the mean of three replicates. Differences were significant at P < 0.05 (*) and P < 0.01 (**) according to Duncan’s multiple range test.

ROS and NO in the plant response to cadmium

The detection in vivo of ROS and NO in roots was carried out using fluorescence probes and was examined in either a fluorescence or confocal laser microscope (Fig. 5). To follow the H2O2 production, DCF-DA was used. To observe O2.− production, DHE was used as a probe, which is specific for this anion (Zhao et al. 2003). The specificity of both probes was checked by using specific ROS scavengers, TMP for O2.−, and ASC for H2O2. As shown in Fig. 5, cadmium induced the over-accumulation of O2.− (Fig. 5a & d) and H2O2 (Fig. 5b & e) in roots. The incubation of roots with different inhibitors showed that O2.− production was inhibited by LaCl3 and therefore was dependent on Ca2+ channels, but phosphatases were not involved because cantharidin did not reduce the fluorescence (Fig. 5a). The O2.− production was depleted by NaN3 and DPI, which suggest that peroxidases and NADPH oxidases are involved in ROS production (Fig. 5a). In lateral roots, O2.− accumulation was mainly located in phloem and cortex cells (Fig. 5a). Analysis of cross-sections of principal roots treated with Cd by confocal laser microscopy showed a higher O2.− accumulation in cell walls from xylem vessels and pericycle and, to a lower extent, in the epidermis and cortex (Fig. 5d). H2O2 followed a similar pattern in response to Cd in both lateral (Fig. 5b) and principal roots (Fig. 5e). Cadmium induced H2O2 generation in lateral roots, mainly in the cell wall of epidermis, phloem and xylem cells, but also in cortex cells (Fig. 5b). Cross-sections of principal roots showed similar results to those observed for O2.− radicals (Fig. 5e). In both cases, fluorescence was mainly detected in xylem vessels where it was distributed in a Y pattern.

Figure 5.

Imaging of reactive oxygen species (ROS) (O2·−, H2O2) and nitric oxide (NO) production in pea roots. (a) O2·−-dependent DHE fluorescence in lateral roots from control and Cd-treated pea plants. As negative control, roots were incubated with 1 mm TMP, an O2 scavenger. Root pieces of 2 cm from the cap were incubated with different inhibitors before DHE labelling: 5 µm cantharidin (cant, protein phosphatase inhibitor), 1 mm LaCl3 (Ca2+ channel blocker), 10 µm DPI (NADPH oxidase inhibitor), 1 mm N3Na (peroxidase inhibitor). Panel on the right corresponds to cross-sections obtained at 6 mm to the cap. (b) H2O2-dependent DCF-DA fluorescence in lateral roots from control and Cd-treated pea plants. As negative control, roots were incubated with 1 mm ascorbate (ASC), which acts as an H2O2 scavenger. Panel on the right corresponds to cross-sections obtained at 6 mm from the cap. (c) NO-dependent DAF-2DA fluorescence in lateral roots from control and Cd-treated pea plants. As negative control, roots were incubated with 1 mm aminoguanidine (AG), which is a mammalian NOS inhibitor. Panel on the right shows cross-sections obtained at 6 mm from the cap. (d) O2·−-dependent DHE fluorescence in principal roots from control and Cd-treated pea plants. Panel on the right shows a magnification of the vascular cylinder. (e) H2O2-dependent DCF-DA fluorescence in principal roots from control and Cd-treated pea plants. Panel on the right shows a magnification of the vascular cylinder. (f) NO-dependent DAF-2 DA fluorescence in principal roots from control and Cd-treated pea plants. Panel on the right shows a magnification of the vascular cylinder.

The analysis of NO accumulation in principal roots with DAF-2 DA showed opposite results to those observed for ROS production, with a strong decrease of NO-dependent fluorescence by Cd treatment (Fig. 5c). The maximum fluorescence in control roots was observed in the caps and was abolished by aminoguanidine, a NOS inhibitor (Fig. 5c). In cross-sections of lateral roots from control plants, the production of NO was mainly observed in the epidermis, and in principal roots it was detected in the cortex, xylem and, to a lower extent, in phloem (Fig. 5f). To study the possible cell death induced by cadmium, roots were incubated with Evan’s Blue and a strong staining was observed in Cd-treated roots, which is indicative of a loss of cell integrity and possibly cell death (Fig. 6).

Figure 6.

Cadmium effect on cell death in pea roots. Viability of cells was visualized by Evan’s Blue staining as described in the Materials and Methods section. C, control roots; Cd2+, Cd-treated roots.

SA, JA and ET levels under cadmium stress

To get more information on the mechanism involved in cell response to Cd toxicity, levels of JA and SA, and ET were determined by GC-MS and GC, respectively. As shown in Fig. 7, cadmium induced a twofold increase in the total content of SA and JA, the conjugated SA and JA (MeSA and MeJA) being the main forms detected in both control and Cd-treated roots. Similar results were obtained for ET, although the Cd-induced increase of ET was smaller than that observed for SA and JA.

Figure 7.

Jasmonic acid (JA) and salicylic acid (SA) production, and ethylene (ET) emission in roots from control and Cd-treated pea plants. JA and SA contents were determined by gas chromatography (GC)-mass spectrometry, and ET was assayed by GC as described in the Materials and Methods section. Each rectangle represents the mean ± SE of 12 replicates. Differences were significant at P < 0.05 (*), P < 0.01 (**) and P < 0.001 (***) according to Duncan’s multiple range test. MeJA, methyl-jasmonate; MeSA, methyl-salicylate; FW, fresh weight.


Cadmium inhibits root growth and induces morphological and ultrastructural modifications

In a previous work carried out in our laboratory, we demonstrated that pea plant grown with 50 µm CdCl2 accumulated Cd mainly in the roots (4.44 mg Cd g−1 dry weight) whose growth was 1.36-fold reduced (Sandalio et al. 2001). As shown in this work, the Cd-dependent reduction of root growth was mainly attributed to the reduction in the number and length of lateral roots. The ultrastructural analysis of root cross-sections showed changes in the vascular cylinders of principal roots, which affected mainly the xylem vessels but also the phloem, and could be attributed to either Cd-dependent interferences with cambium differentiation and cell division or to Cd-induced imbalances in phytohormones, such as abscisic (ABA), as it was demonstrated by Barceló, Vázquez & Poschenrieder (1988) in bush bean plants.

Cadmium induces oxidative stress in pea roots

Oxidative stress appears to be involved in Cd toxicity to judge by the decrease in some antioxidant levels and the increase of ROS (O2.−, H2O2) accumulation. Thus, CAT activity was slightly down-regulated at transcriptional level. The decrease of CAT activity by Cd ions was also described in pea (Dixit et al. 2001) and pine (Schützendübel et al. 2001) roots, as well as in sunflower leaves (Laspina et al. 2005), rice (Kuo & Kao 2004) and Arabidopsis leaves (Cho & Seo 2004), while the opposite effect was observed in radish roots (Vitória, Lea & Azevedo 2001) and soybean cell cultures (Sobkowiak et al. 2004). GPX was also down-regulated probably by enzyme inhibition by the metal, such as that described in wheat roots (Converso, Fernández & Tomaro 2000). Reduction of GPX activity was also observed in pea leaves under the same experimental conditions (Sandalio et al. 2001), although opposite results were obtained in P. vulgaris leaves (Smeets et al. 2005) and soybean cell cultures (Sobkowiak et al. 2004). GR, a key enzyme of the ASC–GSH cycle (Jiménez et al. 1997; Noctor & Foyer 1998), was down-regulated probably by oxidative modifications of the protein (Romero-Puertas et al. 2002). These results are not in accordance with those reported by Dixit et al. (2001) in the same plant species, and in P. vulgaris leaves (Smeets et al. 2005), rice (Kuo & Kao 2004) or radish (Vitória et al. 2001) where an increase of the GR activity was observed. The reduction of the total GSH content observed in roots from plants exposed to Cd could be attributed to the synthesis of phytochelatins induced by the metal (Howden et al. 1995; Xiang & Oliver 1998; Zhu et al. 1999). In Medicago sativa roots, Cd also produced a reduction of GSH and an increase of phytochelatins (Ortega-Villasante et al. 2005). The reduction in GSH and ASC levels observed in this work could contribute to a decrease in the ASC–GSH cycle efficiency in the roots tissues.

SOD is a key enzyme in the plant antioxidant defences and is encharged of the dismutation of O2.− radicals to H2O2 and O2 (Alscher, Erturk & Heath 2002). Cadmium down-regulated the CuZn-SODs, while Fe-SOD and Mn-SOD were not very affected, despite the up-regulation of Fe-SOD at transcriptional level. The analysis of total SOD activity in root extracts showed a slight increase by cadmium; however, the assay of total SOD activity in crude extracts can be subjected to interferences by compounds with mimicked SOD activity (Sandalio et al. 2001). Reduction of SOD activity induced by Cd was reported in wheat (Milone et al. 2003) and bean plants (Cardinaels et al. 1984), although the opposite effect was observed in sunflower (Laspina et al. 2005), rice (Kuo & Kao 2004) and soybean cell cultures (Sobkowiak et al. 2004).

The discrepancies observed in the antioxidant activities of different plant species under Cd treatment are probably attributed to the metal concentration and period of treatment used in each case. However, the cell response to Cd is also different depending on the species, organ or tissue (Sobkowiak et al. 2004; Benavides, Gallego & Tomaro 2005), and our results could represent the cell adaptation to the metal after a long period of plant exposure.

The reduction observed in some antioxidants probably explained the higher extent of lipid peroxidation and protein carbonyl groups observed in roots from pea plants exposed to cadmium. However, the involvement of lipoxygenase in lipid peroxidation cannot be discarded. Lipid peroxidation dependent on lipoxygenase activity was demonstrated under heavy metal stress in P. vulgaris (Somashekaraiah et al. 1992) and Arabidopsis thaliana (Skórzynska-Polit et al. 2006), and in senescence and wounding in different plant species (Spiteller 2003).

Cadmium induces ROS over-accumulation and reduction of NO content in roots

The analysis of H2O2 in roots extracts and in root tissues in vivo by fluorescence microscopy showed a strong accumulation of this ROS specially in the root tips (Fig. 5). A Cd-dependent accumulation of peroxides was also observed in M. sativa plants after 24 h of treatment (Ortega-Villasante et al. 2005). The H2O2 observed in root tips from control plants (Fig. 5b) could be associated to cell differentiation and elongation (Ogawa, Kanematsu & Asada 1997; Potikha et al. 1999; Córdoba-Pedregosa et al. 2003), while the increase in H2O2 observed in Cd-treated roots could be responsible for oxidative damages. In fact, staining with Evan’s Blue, which is considered a cell death marker, showed a higher dye retention in the zone above the root tip (Fig. 6). Similar results were obtained with DHE, the fluorescence probe of O2.− radicals. Concerning to the source of O2.− in Cd-treated roots, the abolishment of fluorescence by NaN3 and DPI points to the involvement of peroxidases and NADPH oxidases in O2.− generation. In spinach hypocotyls and Zinnia elegans stems, the main source of ROS was identified as an NADPH oxidase (Ogawa et al. 1997; Ros Barceló 1999). However, cell wall peroxidases are also important sources of ROS in lignifying tissues and participate in the oxidative burst induced by biotic stress (Neill et al. 2002b; Blee et al. 2003). The Cd-dependent production of O2.− was also abolished by LaCl3, which demonstrates that changes in cytosolic Ca2+ are necessary for the Cd-dependent O2.− accumulation. Conversely, phosphatases are not apparently involved because cantharidin did not remove the fluorescence but only reduced it. These results are in contrast with those obtained in pea leaves under cadmium toxicity, where O2.− production was abolished by cantharidin (Romero-Puertas et al. 2004). This fact evidences the existence of differences between roots and leaves in the regulation of ROS production. An organ-specific transduction network for regulating the effects of ABA and JA on gene expression upon wounding was reported (Dammann, Rojo & Sánchez-Serrano 1997).

Detection of ROS in cross-sections of Cd-treated roots by confocal laser microscopy (CLM) showed the fluorescence mainly located in the cell wall of the xylem vessels, pericycle and epidermis of principal roots, and also in the cell wall of cortex cells in lateral roots. ROS production was associated with lignification of cell wall in xylem (Ogawa et al. 1997; Ros Barceló 1998, 2005). The higher accumulation of ROS in xylem vessels under Cd treatment could be attributed to an accelerated apoptosis in order to produce functional xylem elements. However, aside from its role in lignification, ROS can also play a regulatory role acting as signal molecules to induce cellular responses in neighbouring cells.

The study of NO accumulation by using DAF-2 DA and fluorescence microscopy showed green fluorescence in control lateral root tips, which was strongly reduced by cadmium. However, Bartha, Kolbert & Erdei (2005) observed a Cd-dependent increase of NO accumulation in roots from pea seedlings after 24 h of treatment with the metal. This discrepancy could be attributed to differences in the cell response to short and long periods of metal treatment. The NO detected in control pea plants appears to be mainly produced by NOS activity, to judge by its inhibition by aminoguanidine, a characteristic NOS inhibitor. An inducible AtNOS protein was demonstrated in Arabidopsis roots (Guo et al. 2003). The analysis of root cross-sections showed that in control roots, production of NO occurs in the same places were ROS are generated. Under Cd treatment, the fluorescence attributed to cell walls of cortex cells was considerably reduced, but the fluorescence of xylem did not change. The high production of NO in xylem cells could be a result of apoptosis processes that take place in these lignifying cells (Gabaldón et al. 2005) and the reduction of NO observed in cortex cells is in agreement with the NO decrease reported in senescent pea leaves (Corpas et al. 2004). This suggests that Cd could induce an accelerated senescence in the root tissue, which is supported by the increase observed in the ET level (Leshem et al. 1998). However, neither the mechanism of NO depletion by Cd nor the nature of the protein involved in NO generation are known.

NO was considered by some authors as an antioxidant by its ability to combine with O2.− and prevent oxidative damages (Romero-Puertas & Delledonne 2003), and the production of NO in different biotic and abiotic stresses was demonstrated (Gould et al. 2003). The accumulation of O2.− and depletion of NO by cadmium stress, together with the oxidative damages produced, could support the idea of an antioxidative role of NO. NO was found to alleviate the deleterious effect of Cd on lupine roots (Kopyra & Gwózdz 2003) and sunflower leaves (Laspina et al. 2005). However, NO is also a signal molecule involved in triggering the defence responses of cells in different stress conditions (Neill et al. 2003; Lamattina et al. 2003; Romero-Puertas & Delledonne 2003). Imbalances of the content of O2.− and NO in the tissue could interfere with the signal transduction pathway disturbing the defence mechanisms against stress (Delledonne et al. 2001). NO is also involved in development, especially in promoting lateral root development (Correa-Aragunde, Gaziano & Lamattina 2004), which could explain the strong reduction of lateral roots observed in plants exposed to cadmium.

JA, ET and SA are up-regulated by Cd

To get deeper insights into the mechanisms of cell response to cadmium toxicity, the analysis of JA, SA and ET contents was carried out. All these compounds were induced by Cd treatment, which suggests that they are involved in cell response to Cd toxicity. The rise of ET is in accordance with the Cd-induced senescence previously described in pea leaves (Sandalio et al. 2001; McCarthy et al. 2001) and also with the reduction of NO levels reported in senescent plants (Leshem et al. 1998; Corpas et al. 2004).

JA is an oxylipin that acts as intercellular and intracellular signalling compound in different plant defence situations, such as response to pathogens and herbivore attack (Turner, Ellis & Devoto 2002; Mithöfer, Schulze & Boland 2004). However, responses mediated by JA can be also triggered by diverse abiotic stresses (Turner et al. 2002; Devoto & Turner 2003). JA is synthesized from linolenic acid and its production is associated with lipid peroxidation and membrane damages (Turner et al. 2002; Mithöfer et al. 2004). In this work, we have demonstrated that cadmium induced ROS accumulation and lipid peroxidation in pea roots, which could explain the increase observed in JA production. Oxidative stress is a common point in different situations involving JA, such as pathogen attack, wounding or heavy metal stress. However, the involvement of JA in metal-induced gene expression is not clear (Xiang & Oliver 1998). JA induced the up-regulation of genes for the synthesis of GSH and phytochelatins and providing protection under Cd toxicity (Xiang & Oliver 1998).

SA was proposed to have beneficial effects in different stress conditions such as ultraviolet, heat, salinity, drought and also cadmium (Yalpani, Enyedi & León 1994; Janda et al. 1999; Mishra & Choudhuri 1999; Metwally et al. 2003). In this respect, the enhanced SA content detected in roots could be involved in the cell response to Cd-induced damages. However, the mechanisms involved in the beneficial effects of SA in plants under Cd stress are not completely understood. Metwally et al. (2003) proposed that SA could alleviate Cd toxicity by enhancing repair processes.

In conclusion, in pea plants, long-term exposure to Cd produces oxidative stress in roots as a result of disturbances in antioxidant defences, both enzymatic and non-enzymatic, bringing about an increase in ROS accumulation and decrease in the NO level. The results obtained in this work suggest that enhancement of both the plant antioxidant defences and NO-generating activity should be taken into account to design molecular strategies to improve the tolerance of plants to heavy metals.


M. Rodríguez-Serrano and M.C. Romero-Puertas acknowledge fellowships from the Junta de Andalucía and DGESIC (Ministry of Education and Science, Spain). Special thanks are given to Dr A. Pulido and Dr D. Garrido for their technical assistance in fluorescence microscopy and GC, respectively, and Ms D. Pazmiño, Miss N. de la casa and Mr C. Ruiz for their skilful technical assistance. The laser confocal microscopy analyses were carried out at the Technical Services of the University of Jaén. This work was supported by grant BFI2002-04440-CO2-01 from the DGESIC (Ministry of Education and Culture), and Junta de Andalucía (Research Group CVI 0192), Spain.