Nitric oxide promotes MPK6-mediated caspase-3-like activation in cadmium-induced Arabidopsis thaliana programmed cell death

Authors

  • YUN YE,

    1. MOE Key Laboratory of Laser Life Science & Institute of Laser Life Science, College of Biophotonics, South China Normal University, Guangzhou 510631, China
    Search for more papers by this author
  • ZHE LI,

    1. MOE Key Laboratory of Laser Life Science & Institute of Laser Life Science, College of Biophotonics, South China Normal University, Guangzhou 510631, China
    Search for more papers by this author
  • DA XING

    Corresponding author
    1. MOE Key Laboratory of Laser Life Science & Institute of Laser Life Science, College of Biophotonics, South China Normal University, Guangzhou 510631, China
    Search for more papers by this author

D. Xing. Fax: +86 20 8521 6052; e-mail: xingda@scnu.edu.cn

ABSTRACT

Nitric oxide (NO), a vital cell-signalling molecule, has been reported to regulate toxic metal responses in plants. This work investigated the effects of NO and the relationship between NO and mitogen-activated protein kinase (MAPK) in Arabidopsis (Arabidopsis thaliana) programmed cell death (PCD) induced by cadmium (Cd2+) exposure. With fluorescence resonance energy transfer (FRET) analysis, caspase-3-like protease activation was detected after Cd2+ treatment. This was further confirmed with a caspase-3 substrate assay. Cd2+-induced caspase-3-like activity was inhibited in the presence of the NO-specific scavenger 2-(4-carboxyphenyl)-4,4,5,5-tetramethylimidazoline-1-oxyl-3-oxide (cPTIO), suggesting that NO mediated caspase-3-like protease activation under Cd2+ stress conditions. Pretreatment with cPTIO effectively inhibited Cd2+-induced MAPK activation, indicating that NO also affected the MAPK pathway. Interestingly, Cd2+-induced caspase-3-like activity was significantly suppressed in the mpk6 mutant, suggesting that MPK6 was required for caspase-3-like protease activation. To our knowledge, this is the first demonstration that NO promotes Cd2+-induced Arabidopsis PCD by promoting MPK6-mediated caspase-3-like activation.

INTRODUCTION

Cadmium (Cd2+), a non-essential element, exhibits high levels of toxicity. Its presence as a pollutant is mainly due to industrial activities. It is rapidly taken up by plant roots and transferred into the food chain, posing a significant risk to human health (Pinto et al. 2004; Järup & Akesson 2009). Upon Cd2+ exposure, seed germination and plant growth are inhibited due to alterations in the photosynthesis rate, nutrient uptake and distribution (Sanità di Toppi & Gabbrielli 1999; Sandalio et al. 2001).

Programmed cell death (PCD) is a fundamental process in plants. It is involved in defence, development and stress response (Pennell & Lamb 1997; Reape & McCabe 2008). Depending on the concentration, Cd2+ can trigger either necrosis or PCD in Allium cepa Linnaeus root tip cells (Behboodi & Samadi 2004). It also induces PCD in BY-2 suspension cultures by causing genotoxicity (Fojtová & Kovarík 2000; Fojtováet al. 2002). De Michele et al. (2009) showed that treatments of 100 and 150 µm Cd2+ in Arabidopsis suspension cultures led to a type of PCD that resembles an accelerated senescence process. While well known as primary mechanism in animal apoptosis, (Cohen 1997), caspase activity has also been identified in plants through the use of caspase-specific inhibitors or synthetic fluorogenic caspase substrates (Danon et al. 2004). Caspase-3, the key component of apoptosis, is considered to be the final step executer in many apoptosis pathways (Cohen 1997; Porter & Jänicke 1999). Our recent work (Zhang et al. 2009; Li & Xing 2011) utilized fluorescence resonance energy transfer (FRET) technique to successfully detect caspase-3-like activation in vivo during plant PCD, but the mechanism of regulating caspase-3-like activation in plant PCD remains largely unknown.

Nitric oxide (NO) is an important signalling molecule in plants. It participates in many developmental and physiological processes, including defence responses, PCD, seed germination, lateral root initiation, flowering and stomatal closure (Lamattina et al. 2003; Libourel et al. 2006; Besson-Bard, Pugin & Wendehenne 2008). Previous work has shown that Cd2+ induces NO production, but the reports of exogenous and endogenous NO in Cd2+ stress are conflicting. NO has been demonstrated to act as an antioxidant function and pretreatment with an NO donor scavenges active oxygen species and enhances tolerance to Cd2+ stress in plants (Laspina et al. 2005; Singh et al. 2008). These studies were mainly investigations regarding the effects of exogenous NO. Endogenous NO production, however, is likely to contribute to Cd2+ toxicity in plants. NO formation is related to Cd2+-induced root growth inhibition (Groppa et al. 2008; Besson-Bard et al. 2009). Experiments that have increased reactive oxygen species (ROS) production and SAG12 expression have shown that NO is required for Cd2+-induced cell death in Arabidopsis suspension cultures (De Michele et al. 2009). NO has also been revealed to contribute to Cd2+ toxicity by promoting Cd2+ accumulation (Besson-Bard et al. 2009; Ma et al. 2010). The mechanism of NO function in Cd2+-induced PCD, however, still remains elusive, and research investigating the relationship between NO and caspase-like proteases could shed light on the function of the NO signalling pathway during plant PCD.

The mitogen-activated protein kinase (MAPK) cascade plays a vital role in intracellular signal transduction. As shown in previous studies, MAPKs are not only involved in biotic and abiotic signalling, but also in processes such as hormonal and developmental signalling (Tena et al. 2001; Zhang & Klessig 2001; Colcombet & Hirt 2008). Cd2+-induced MAPK pathways have been demonstrated in alfalfa (Jonak, Nakagami & Hirt 2004) and rice (Yeh, Chien & Huang 2007), but the molecular mechanisms of MAPK are far from clear. It has been demonstrated that NO is involved in activating MAPK signalling pathways. NO mediates indole acetic acid-induced MAPK activation during the adventitious rooting process in cucumber (Cucumis sativus) and abscisic acid-induced MAPK activation in Arabidopsis (Pagnussat et al. 2004; Zhang et al. 2007). Based on these studies, we hypothesize that there may be some relationship between NO and MAPK in Cd2+-induced PCD.

The aim of this research is to explore a potential NO signalling pathway in Arabidopsis thaliana under ubiquitous toxic heavy metal Cd2+ exposure. We monitored caspase-3-like activation and NO production in vivo after Cd2+ treatment. It was observed that Cd2+-induced endogenous NO production contributes to caspase-3-like activation. By investigating potential NO signalling pathways, we found that NO promoted Cd2+-induced MPK6 activation and caspase-3-like activation that ultimately led to PCD. These results clearly have shown that NO promotes MPK6-mediated caspase-3-like activation and is an important regulator in Cd2+-induced Arabidopsis PCD.

MATERIALS AND METHODS

Plant materials and growth conditions and treatments

The seeds of ECFP-DEVD-EYFP transgenic Arabidopsis were kindly provided by Dr Shengjun Nie of our group. In the present experiments, A. thaliana (ecotype Columbia) wild type (WT), T-DNA insertion mutants of mpk3-1 (SALK_151594), mpk4-2 (SALK_056245), mpk6-2 (SALK_073907) and mpk6-3 (SALK_127507) were used. Seeds of Arabidopsis were surface sterilized and grown on solid Murashige and Shoog (MS) medium as described previously (Li, Yue & Xing 2012). Then seedlings were transferred to liquid MS medium containing different concentrations of Cd2+, sodium nitroprusside (SNP) and cPTIO for the indicated time points.

Cell viability assay

To determine cell viability, at the indicated time points following the treatments, 14-day-old seedlings were incubated with 50 µm fluorescein diacetate (FDA) for 10 min at room temperature in darkness and rinsed twice in ultrapure water. The fluorescence of FDA was observed under a Zeiss LSM 510 (Jena, Germany). The intensity of FDA fluorescence in root tips was quantified based on an auto microplate reader (infinite M200, Tecan, Grödig, Austria) using an excitation wavelength of 488 nm and an emission wavelength of 521 nm. Twenty milligram root strips (5∼10 mm) were used for each measurement.

Hoechst 33258 staining

After being treated as described previously, 14-day-old seedlings were collected, washed twice with ultrapure water and then stained with Hoechst 33258 staining solution according to the manufacturer's instructions (Beyotime, Haimen, China). They were observed under a Nikon fluorescence microscope (Nikon USA, Melville, NY, USA) attached to a hydrargyrum lamp with an excitation filter of 330–380 nm and emission filter of 450–490 nm.

Fluorometric assay for caspase-3-like activity

Caspase-3-like activity was measured using the fluorogenic substrates Ac-DEVD-AFC according to the manufacturer's instructions (Alexis Biochemicals, Lausen, Switzerland). At the indicated time points, 0.4 g seedlings were ground to powder in liquid nitrogen. Then the samples were resuspended in assay buffer [20% glycerol, 0.1% Triton, 10 mm ethylenediaminetetraacetic acid (EDTA), 3 mm dithiothreitol (DTT), 2 mm phenylmethylsulfonyl fluoride (PMSF) and 50 mm sodium acetate, pH 7.4] and incubated with 100 µm Ac-DEVD-AFC. Caspase-3-like activity was measured continuously by monitoring the release of fluorogenic AFC at 37 °C. The excitation wavelength of AFC was 400 nm and the emission wavelength was 505 nm as determined by an auto microplate reader (infinite M200, Tecan).

Laser confocal scanning microscopy and FRET analysis

Confocal imaging and FRET analysis were performed on a Zeiss LSM 510 laser confocal scanning microscope (LCSM). The specific settings of light excitation and emissions used are as follows: FDA and DAF-FM DA (Ar ion laser, Ex. 488 nm, Em. BP 500–550 nm); MitoTracker Red CMXRos (He-Ne laser, Ex. 543 nm, Em. BP 565–615 nm); and chloroplast autofluorescence (Ar ion laser, Ex. 488 nm, Em. LP 650 nm). The quantitative analysis of the fluorescence images was performed with Zeiss Rel 4.2 image processing software.

We thank Dr Shengjun Nie of our group for the ECFP-DEVD-EGFP transgene construction and plant screening (unpublished observation). The fusion ECFP-DEVD-EYFP proteins contain an 18-amino acid linker peptide (SSSELSGDEVDGTSGSEF) possessing the caspase-3 cleavage sequence DEVD, which joins the C terminus of ECFP to the N terminus of EYFP. The activation of caspase-3 leads to the cleavage of the linker peptide, which then results in a conformational change that decreases the FRET ratio (Zhang et al. 2009; Li & Xing 2011). For FRET detection, the 458 nm laser line from an Ar ion laser was used for ECFP excitation. Because EYFP is not excited at 458 nm, the presence of EYFP fluorescence indicated the occurrence of FRET between ECFP and EYFP. Two band-pass (BP) filters (BP 465–510 nm and BP 520–550 nm) were used for ECFP and EYFP, respectively, and the chlorophyll autofluorescence was collected with a 650 to 700 nm BP filter. All of the quantitative analyses of the fluorescent images were performed using Zeiss Rel4.2 image-processing software.

NO live imaging and measurement

DAF-FM DA is a membrane-permeable fluorogenic reagent useful in real-time NO detection because of its ability to form the fluorescent benzotriazole derivative DA F-FM-T (Balcerczyk, Soszynski & Bartosz 2005). Although other non-specific reactions may interfere with the probe specificity (Balcerczyk et al. 2005; Planchet & Kaiser 2006), these effects can be accounted for by the use of suitable controls such as the NO scavenger cPTIO (Wilkins et al. 2011).

For NO detection in seedlings, 14-day-old seedlings were collected after being treated, incubated with 20 µm DAF-FM DA in 10 mm Tris-HCl (pH 7.4) for 2 h and then washed three times with W5 solution. The fluorescence was visualized using a confocal laser scanning microscope (Zeiss LSM 510). The fluorescence intensity of DAF in roots was detected with a LS 55 Luminescence Spectrophotometer (PerkinElmer, LS55, Beaconsfield, UK) set for an excitation wavelength of 488 nm and emission wavelengths between 500 and 550 nm. The fluorescence intensity at 515 nm was used to determine the relative NO generation. Twenty milligram root strips (5∼10 mm) were used for each measurement.

For intracellular NO detection, Arabidopsis protoplasts were double stained with 1 µm DAF-FM DA and 0.2 µm MitoTracker Red CMXRos for 30 min in the dark and fluorescence was determined as described previously (Yao & Greenberg 2006).

Haemoglobin analysis

The NO content was also determined by the haemoglobin method as described previously (Murphy & Noack 1994). One gram each of roots from treated seedlings was incubated with 100 units of catalase (CAT) and superoxide dismutase (SOD) for 5 min to remove endogenous ROS. Then, 10 mL of 5 mm oxyhaemoglobin was added and incubated for 10 min incubation, and the NO content was measured spectrophotometrically by measuring the conversion of oxyhaemoglobin to methemoglobin.

Protein extraction and Western blot analysis

For protein extraction, 0.4 g treated 14-day-old seedlings were ground to powder in liquid nitrogen and dissolved in extraction buffer [50 m mm Tris–HCl, pH 6.8, 50 m mm DTT, 4% (w/v) SDS, 10% (v/v) glycerol, 1% (w/v) polyvinylpolypyrrolidone (PVPP), 5 m mm PMSF]. After centrifugation at 12 000 g for 20 min twice at 4 °C, the supernatants were transferred into new tubes, quickly stored at −80 °C until analysed. The protein content was determined by the Bradford (1976) method. Extracts containing 20 µg protein were separated by 12% sodium dodecyl sulphate–polyacrylamide gel electrophoresis (SDS–PAGE) and transferred to nitrocellulose membranes. The membranes were blocked with TBST (10 mm Tris–HCl, pH 7.4, 150 mm NaCl, 0.1% Tween-20) containing 5% non-fat milk for 1 h and then incubated with phosphoP44/42 MAPK antibody (Cell Signaling Technology, Beverly, MA, USA) (Wang et al. 2010). After washing three times, the membranes were blocked with secondary antibody anti-rabbit IRDyeTM800 (Rockland Immunochemicals, Gilbertsville, PA, USA). Proteins were detected by using Odyssey two-color infrared imaging system (Li-Cor, Inc., Lincoln, NE, USA).

RESULTS

Activation of caspase-3-like protease induced by Cd2+

In order to explore the mechanism of Cd2+-induced PCD, the effects of Cd2+ on caspase-3-like activation in vivo were monitored by FRET in protoplasts of ECFP-DEVD-EYFP transgenic Arabidopsis. With or without 100 µm Cd2+ exposure, the real-time ECFP and EYFP fluorescence images were collected with LSM microscopy (Fig. 1a,c), and the fluorescence intensities of images and their ratios (EYFP/ECFP) were shown in Fig. 1b,d. After Cd2+ exposure for 16.5 h, the EYFP/ECFP FRET ratio started to decrease compared with samples without Cd2+ treatment (Fig. 1c,d), suggesting that the activation of caspase-3-like protease was causing the cleavage of the fusion protein ECFP-DEVD-EYFP.

Figure 1.

Activation of caspase-3-like protease during Cd2+ exposure. Transgenic ECFP-DEVD-EYFP Arabidopsis protoplasts without (a,b) or with (c,d) Cd2+ treatment were excited by a 458 nm laser. ECFP and EYFP images were recorded, and the fluorescence intensity of ECFP, EYFP and chlorophyll, as well as the EYFP/ECFP ratio were measured. Scale bars = 10 µm. (e) 14-day-old WT seedlings were treated with 100 µm Cd2+ for the indicated time. The detection of caspase-3 activation by fluorogenic substrate Ac-DEVD-AFC is described in Materials and Methods. (f) Detection of caspase-3-like activity after the 14-day-old WT seedlings were transferred to hydroponic MS medium for 2 or 3 d. The different treatments were as follows: control, 100 µm Cd2+, 100 µm Cd2++ 100 µm Z-VAD-FMK, 100 µm Cd2++ 100 µm Ac-DEVD-CHO, 100 µm Cd2++ 0.5% DMSO. The data are the mean ± SE from three independent experiments, and the asterisks represent differences with P < 0.01 (**) versus the control group. a.u., arbitrary units; DIC, differential interference contrast; MS, Murashige and Skoog.

Furthermore, the caspase-3-like activity in WT seedlings was detected by fluorogenic substrate Ac-DEVD-AFC (100 µm). Fourteen-day-old WT Arabidopsis plants were treated with 100 µm Cd2+ for 0–7 d before their caspase-3-like activity was tested in extracts. Compared with the controls, Cd2+ induced a nearly 2.5-fold increase of caspase-3-like activity (Fig. 1e). Inhibitor analysis by FRET and the fluorogenic substrate Ac-DEVD-AFC showed that the addition of mammalian general caspase inhibitor Z-VAD-FMK and caspase-3 inhibitor Ac-DEVD-CHO blocked decreases in EYFP fluorescence, the EYFP/ECFP ratio and caspase-3-like activity (Fig. 1f & Supporting Information Fig. S1). All these data confirmed that Cd2+ induced the activation of caspase-3-like protease.

Caspase-3-like protease is required for Cd2+-induced PCD

To further confirm that Cd2+-induced PCD in Arabidopsis is correlated with caspase-3-like activity, Arabidopsis WT seeds were germinated and grown for 10 d in different vertical MS plates. In the presence of Z-VAD-FMK and Ac-DEVD-CHO, plant growth was much better than those grown with Cd2+ alone (Fig. 2a).

Figure 2.

Effects of Z-VAD-FMK and Ac-DEVD-CHO on Cd2+-induced PCD. (a) WT seedling were grown in MS medium containing 100 µm Cd2+ and supplemented with 100 µm Z-VAD-FMK or 100 µm Ac-DEVD-CHO. Pictures of seedlings were taken 10 d after germination. For root length analysis, 10 seedlings were measured and averaged. The data are the mean ± SE from three independent experiments. (b) 14-day-old WT seedlings were transferred to hydroponic MS medium for 3 d. The different treatments were control (without treatment), 100 µm Cd2+, 100 µm Cd2++ 100 µm Z-VAD-FMK and 100 µm Cd2++ 100 µm Ac-DEVD-CHO. Then seedlings were subjected to 50 µm FDA staining and root fluorescence images were analysed by LCSM. Scale bars = 50 µm. (c) The FDA fluorescence of viable root cells under the indicated treatment was calculated. The different treatments were control (without treatment), 100 µm Z-VAD-FMK, 100 µm Ac-DEVD-CHO, 0.5% DMSO, 100 µm Cd2+, 100 µm Cd2++ 100 µm Z-VAD-FMK, 100 µm Cd2++ 100 µm Ac-DEVD-CHO and 100 µm Cd2++ 0.5% DMSO. In the figure, each value is the mean ± SE of five replicates and the asterisks indicate a difference with P < 0.01 (**), versus the control group. (d) 14-day-old WT seedlings were transferred to hydroponic MS for 3 d with the indicated treatments as described in (c). The seedlings were then subjected to Hoechst 33258 staining. Fluorescent images of the roots were collected via a Nikon fluorescence microscope with an attached hydrargyrum lamp. The points with bright blue fluorescence indicate condensed chromatin. a.u., arbitrary units; LCSM, laser confocal scanning microscope; MS, Murashige and Skoog; PCD, programmed cell death; WT, wild type.

The negative effect of caspase inhibitor on Cd2+ toxicity was further demonstrated by FDA staining experiments. As shown in the Fig. 2b confocal images, the viable root cells could be stained by FDA. Quantitative analysis of FDA fluorescence was shown in Fig. 2c. Exposure to Cd2+ for 3 d resulted in a significant loss of FDA fluorescence. Furthermore, the addition of 100 µm of the caspase inhibitors Z-VAD-FMK or Ac-DEVD-CHO caused an increase in cell viability in root cells. Similar results were obtained in leaves of 14-day-old seedlings by FDA staining (Supporting Information Fig. S7).

To evaluate the status of cell death directly, morphological examination was performed with Hoechst 33258. Hoechst 33258 stains the condensed chromatin in apoptotic cells more brightly than normal chromatin because of enhanced membrane permeability in apoptotic cells. As shown in Fig. 2d & Supporting Information Fig. S8, points with bright blue fluorescence were clearly observed in roots and leaves after the 14-day-old seedlings were exposed to 100 µm Cd2+ for 3 d. The control group exhibited a normal appearance containing uniformly stained chromatin with light blue fluorescence. The addition of Z-VAD-FMK or Ac-DEVD-CHO caused a decrease in the number of bright blue fluorescence spots when compared with the Cd2+-treated group. These observations indicated that Cd2+-induced PCD in Arabidopsis correlated with caspase-3-like activity.

NO accumulation in Cd2+-induced PCD

Cd2+-induced NO production in Arabidopsis was investigated with the NO-sensitive fluorescent probe DAF-FM DA. Compared with the control plants, 24 h Cd2+ treatment induced an increase in DAF fluorescence in the roots and leaves of 14-day-old seedlings, and this effect was much stronger after 48 h treatments. This Cd2+-induced DAF fluorescence increase was greatly suppressed by the NO scavenger cPTIO (Fig. 3a & Supporting Information Fig. S6). Relative DAF fluorescence densities in the roots were shown in Fig. 3b, and the data revealed that the effect of Cd2+ on NO production was time dependent. Additionally, there was a significant difference in relative DAF fluorescence densities between samples treated with 100 µm Cd2+ for 6 h and a control group lacking this treatment. After Cd2+ treatment for 24 and 48 h, the fluorescence densities of DAF were approximately two- and threefold higher than that of the control group. By contrast, there was no obvious difference between the control group and the cPTIO additional group. A similar trend in NO production under 100 µm Cd2+ treatment was shown by haemoglobin analysis (Fig. 3c). The NO scavenger cPTIO can eliminate NO by converting it into nitrite (Planchet & Kaiser 2006). Therefore, the nitrite content in the 14-day-old WT seedlings was examined by Griess assay at the indicated time. In Supporting Information Fig. S3, 100 µm Cd2+ could induce nitrite production, and supplementing cPTIO resulted in higher nitrite production.

Figure 3.

Accumulation of NO during Cd2+ induction. (a) 14-day-old WT seedlings were treated with 100 µm Cd2+ for 24 or 48 h, 100 µm Cd2++ 0.5 mM cPTIO for 48 h, or no treatment (control group) for 48 h. The seedlings were then loaded with 20 µm DAF-FM DA for 2 h. The accumulation of NO in the roots was detected by LCSM. Scale bars = 50 µm. (b) At the indicated time, the DAF fluorescence densities in 20 mg roots were detected by LS 55 Luminescence Spectrophotometer. (c) NO contents in the roots were examined by haemoglobin analysis at the indicated time. Data are means ± SE of three experiments, and the asterisks indicate differences with P < 0.05 (*) or P < 0.01 (**) versus the control group. a.u., arbitrary units; FW, fresh weight; LCSM, laser confocal scanning microscope; NO, nitric oxide.

Intracellular NO was detected by double staining with 1 µm DAF-FM DA and 0.2 µm MitoTracker Red CMXRos as described previously (Yao & Greenberg 2006). Firstly, we exposed WT protoplasts to 100 µm Cd2+ for the indicated times. We found that protoplasts exposed to Cd2+ exhibited an increase in fluorescence first in the mitochondrial regions and subsequently in chloroplasts and even the whole cell as the treatment time increased (Supporting Information Fig. S4).

NO contributes to caspase-3-like activation in Cd2+-induced PCD

To assess whether Cd2+-induced NO production correlated with caspase-3-like activity, we assayed caspase-3-like activity in the presence of cPTIO and SNP during Cd2+ treatment. As shown in Fig. 4, at 3 d after Cd2+ treatment, we observed a 2.36 ± 0.14-fold increase of caspase-3-like activity over the control group. The addition of cPTIO reduced this increase to 1.52 ± 0.18-fold. By contrast, SNP increased the Cd2+-induced caspase-3-like activity to 2.63 ± 0.21-fold over the control group. In conclusion, Cd2+-induced caspase-3-like activity was increased by NO production.

Figure 4.

Effects of cPTIO and SNP on caspase-3-like activity induced by Cd2+. 14-day-old Arabidopsis seedlings were transferred to hydroponic MS medium for 2 or 3 d. The different treatments were as follows: control (without treatment), 100 µm Cd2+, 100 µm Cd2++ 1 µm SNP or100 µm Cd2++ 0.5 mm cPTIO. Caspase-3-like activity was detected by the fluorogenic substrate Ac-DEVD-AFC. The data are the mean ± SE of three replicates, and the asterisks indicate P < 0.01 (**) versus the control group. SNP, sodium nitroprusside; MS, Murashige and Skoog.

We then investigated the effect of NO on Cd2+-induced PCD. Arabidopsis WT seeds were germinated and grown for 10 d in different vertical MS plates. The plant growth was much better in the presence of cPTIO than those grown with Cd2+ alone. The addition of SNP, however, inhibited plant growth (Fig. 5a). Furthermore, FDA and Hoechst 33258 stainings were performed in 14-day-old seedlings. As shown in Fig. 5b,c, the addition of cPTIO increased cell viability more than the Cd2+-treated group, whereas the addition of SNP decreased cell viability. Similarly the presence of cPTIO significantly reduced the amount of root cells undergoing PCD, but SNP increased the amount (Fig. 5d). Similar results were found in the leaves of 14-day-old seedlings by FDA and Hoechst 33258 staining (Supporting Information Figs S7 & S8). These results suggested that NO contributed to Cd2+-induced PCD in Arabidopsis through promoting caspase-3-like activation.

Figure 5.

Effects of SNP and cPTIO on Cd2+-induced PCD. (a) Arabidopsis seedlings were grown in MS medium with or without 100 µm Cd2+ and supplemented with 0.5 mm cPTIO or 1 µm SNP. Pictures of seedlings were taken 10 d after germination. For root length analysis, 10 seedlings were measured and averaged. Data are the mean ± SE from three independent experiments. (b) 14-day-old WT seedlings were transferred to hydroponic MS medium for 3 d. The different treatments were control (without treatment), 100 µm Cd2+, 100 µm Cd2++ 1 µm SNP or 100 µm Cd2++ 0.5 mM cPTIO. The seedlings were subjected to 50 µm FDA staining and were analysed by LCSM. Scale bars = 50 µm. (c) The FDA fluorescence of viable root cells under the indicated treatment was calculated. The different treatments were control (without treatment), 0.5 mm cPTIO, 1 µm SNP, 100 µm Cd2+, 100 µm Cd2++ 1 µm SNP or 100 µm Cd2++ 0.5 mm cPTIO. In the figure, each value is the mean ± SE of five replicates, and the asterisks indicate differences with P < 0.01 (**) versus the control group. (d) 14-day-old WT seedlings were transferred to hydroponic MS for 3 d with the indicated treatments as described in (c). The seedlings were then subjected to Hoechst 33258 staining and fluorescent images of the roots were collected via a Nikon fluorescence microscope with an attached hydrargyrum lamp. The points with bright blue fluorescence indicate condensed chromatin. a.u., arbitrary units; LCSM, laser confocal scanning microscope; MS, Murashige and Skoog; PCD, programmed cell death; SNP, sodium nitroprusside; WT, wild type.

NO mediates Cd2+-induced activation of MAPKs

The effect of the NO scavenger cPTIO on MAPK activity in Cd2+-treated Arabidopsis WT seedlings was examined in Fig. 6. After pretreatment with 0.5 mm cPTIO for 1 h, Cd2+-induced activation of 44 and 47 kDa MAPKs were suppressed within 60 min, suggesting that the accumulation of NO was also necessary for the activation of 44 and 47 kDa MAPKs during Cd2+ induction. A quantitative analysis of the levels of phospho-MAPK in seedlings was shown in Fig. 6b.

Figure 6.

Effects of cPTIO on Cd2+-induced activation of MAPKs. (a) WT plants were pretreated for 1 h with or without 0.5 mm cPTIO. Following 100 µm Cd2+ treatment for the time indicated, total extracts were analysed using an in-gel kinase assay. (b) Quantitative analysis of the levels of phospho-MAPK in seedlings that received different treatments. Data represent the mean ± SE, and the asterisks represent statistically significant differences between Cd2+-treated seedlings and cPTIO-pretreated seedlings, P < 0.05 (*), P < 0.01 (**). a.u., arbitrary units; MAPK, mitogen-activated protein kinase; WT, wild type.

MPK6 promotes caspase-3-like activity in Cd2+-induced PCD

A common inhibitor of MAPK kinases, PD98059, was used to clarify whether the MAPK activity was involved in triggering Cd2+-induced caspase-3-like activity. As shown in Fig. 7a, Cd2+-induced caspase-3-like activity was greatly decreased in the presence of PD98059. Using different MAPK mutants, including mpk3-1, mpk4-2 and mpk6-3, we found that only mpk6-3 exhibited a decrease in caspase-3-like activity after exposure to 100 µm Cd2+ for 2 or 3 d. It was suggested that MPK6 might be the main MAPK kinase involved in triggering Cd2+-induced caspase-3-like activity (Fig. 7b). Our conclusion was confirmed by results from the other mpk6 mutant (mpk6-2), showing the same suppressed caspase-3-like activity (Fig. 7c).

Figure 7.

The contribution of MPK6 to Cd2+-induced caspase-3-like activity. (a) 14-day-old WT Arabidopsis seedlings were transferred to MS medium and treated with or without 100 µm Cd2+ or added with 150 µm PD98059 for 2 or 3 d. Caspase-3-like activity was detected by the fluorogenic substrate Ac-DEVD-AFC. Data are the mean ± SE of three replicates. Asterisks represent differences with P < 0.01 (**) versus the control group. (b) 14-day-old WT, mpk3-1, mpk4-2 and mpk6-3 Arabidopsis seedlings were transferred to MS medium and treated with 100 µm Cd2+ for the indicated times. Caspase-3-like activity was detected by the fluorogenic substrate Ac-DEVD-AFC. In the figure, data are the mean ± SE of three replicates, and the asterisks represent differences with P < 0.01 (**) versus the WT group. (c) 14-day-old WT, mpk6-2 and mpk6-3 Arabidopsis seedlings were transferred to MS medium with 100 µm Cd2+ for the indicated times. Data are the mean ± SE of three replicates, and the asterisks represent differences with P < 0.05 (*) or P < 0.01 (**) versus the WT group. MS, Murashige and Skoog; WT, wild type.

In plant growth experiments, plant growth was much better in the mpk6-2 and mpk6-3 mutants than that in WT, mpk3-1 and mpk4-2 seedlings after treatment with 100 µm Cd2+ (Fig. 8a). After 3 d of Cd2+ exposure, both mpk6-3 mutant and the WT seedlings with pretreatment of PD98059 showed higher levels of root cell viability than WT, mpk3-1 or mpk4-2 (Fig. 8b,c). As shown in the Hoechst 33258 staining of roots in Fig. 8d, fewer mpk6 mutant cells underwent PCD than WT cells. Similar results were obtained from FDA and Hoechst 33258 fluorescent imaging in leaves (Supporting Information Figs S7 & S8). Taken together, these results indicate that MPK6 deficiency could suppress Cd2+-induced PCD.

Figure 8.

MPK6 deficiency suppressed cell death induced by Cd2+-treatment. (a) WT, mpk3-1, mpk4-2, mpk6-2 and mpk6-3 seeds were germinated and grown in MS medium containing 100 µm Cd2+. Pictures of seedlings were taken 10 d after germination. For root length analysis, 10 seedlings were measured and then averaged. Each value is the mean ± SE from three independent experiments. (b) Arabidopsis seedlings of WT, mpk3-1, mpk4-2and mpk6-3 were treated with 100 µm Cd2+ for 3d and then incubated with 50 µm FDA. The root cell viability assays were analysed by LCSM. Scale bars = 50 µm. (c) The FDA fluorescence of roots from 14-day-old seedlings with or without 100 µm Cd2+ treatment was calculated. The samples were WT, WT in the presence of 150 µm PD98059, mpk3-1, mpk4-2 and mpk6-3. In the figure, each value is the mean ± SE of five replicates, and the asterisks represent differences with P < 0.01 (**) versus the WT group. (d) Hoechst 33258 staining of root from 14-day-old WT, mpk3-1, mpk4-2, mpk6-2 and mpk6-3 seedlings after 3 d of 100 µm Cd2+-treatment. The points with bright blue fluorescence indicate condensed chromatin. a.u., arbitrary units; LCSM, laser confocal scanning microscope; MS, Murashige and Skoog; WT, wild type.

DISCUSSION

NO regulates many plant physiological processes and acts as a signalling or a toxic molecule depending on the physiological context. Several reports have been published about NO function in Cd2+ stress. Supplementation of NO by SNP significantly reduces Cd2+-induced oxidative damage in wheat roots, indicating an antioxidative role of NO in response to Cd2+ toxicity (Singh et al. 2008). Although several studies have demonstrated that exogenous NO supplements function as antioxidants, Cd2+-induced endogenous NO production is thought to contribute to cadmium toxicity in Arabidopsis seedlings and suspension cultures (Besson-Bard et al. 2009; De Michele et al. 2009), tobacco BY-2 cells (Ma et al. 2010), and wheat roots (Groppa et al. 2008). The mechanism by which NO exerts its effects on Cd2+-induced cytotoxicity is far from clear. In this work, we observed NO accumulation, monitored caspase-3-like activation in vivo and found that caspase-3-like protease was activated by NO and functioned downstream of MPK6 during the Cd2+ stress response. The results of this work provide evidence for the involvement of NO signalling in Cd2+-induced PCD in Arabidopsis.

Caspase-3-like protease is activated during Cd2+-induced Arabidopsis PCD

Previous studies have proved that caspases (Cys-containing Asp-specific proteases) activation plays a central role in apoptosis (Cohen 1997). Caspase-like activity is also involved in various physiological processes in plants (Chichkova et al. 2004; Danon et al. 2004; Vacca et al. 2006). In our recent work, FRET imaging has been successfully used to detect real-time caspase-3-like activation in vivo during UV or Al-induced Arabidopsis protoplasts PCD (Zhang et al. 2009; Li & Xing 2011). In the current study, caspase-3-like activity in protoplasts from ECFP-DEVD-EYFP transgenic Arabidopsis under Cd2+ stress was detected using the FRET technique (Fig. 1c,d). The synthetic fluorogenic substrate Ac-DEVD-AFC has been widely used to measure caspase-3-like activity in Arabidopsis seedlings (Danon et al. 2004). The results also demonstrated the timing of the activation of caspase-3-like protease in Cd2+-stressed seedlings by Ac-DEVD-AFC (Fig. 1e). Inhibitor analysis by FRET and fluorogenic substrate Ac-DEVD-AFC showed that Cd2+-induced caspase-3-like activity was blocked by the mammalian general caspase inhibitor Z-VAD-FMK and the caspase-3 inhibitor Ac-DEVD-CHO (Supporting Information Fig. S1 & Fig. 1f). This result is in agreement with a previous report that CdSO4 induces cell death in tomato suspension cells via a pathway that includes caspase-like proteases (Iakimova et al. 2008). Similarly, during Cd2+ exposure, supplementation with Z-VAD-FMK and Ac-DEVD-CHO promoted root growth and cell viability and also reduced the amount of root cells undergoing PCD (Fig. 2). These results confirmed that caspase-3-like protease was involved in Cd2+-induced Arabidopsis PCD.

NO contributes to Cd2+-induced PCD by facilitating caspase-3-like protease activation

Besson-Bard et al. (2009) have shown a strong NO production in 3-week-old Arabidopsis after 7 h of treatment with 200 µm Cd2+. This process is sensitive to mammalian NOS inhibitors, and while neither AtNOA1 nor NR is involved, the expression of the root iron transporter IRT1 is required. In our study, we found that after treatment with 100 µm Cd2+ for 24 and 48 h, the fluorescence densities of DAF were approximately two- and threefold higher than the control group (Fig. 4). However, 100 µm Cd2+ caused a smaller increase in DAF fluorescence than 200 µm Cd2+ in Besson-Bard et al. (2009). The difference in NO production may be due to the difference in Cd2+ concentration. As shown in Supporting Information Fig. S5, treatment by 200 µm Cd2+ for 7 h induced much stronger DAF fluorescence in 3-week-old Arabidopsis roots than 100 µm Cd2+. Therefore, the effects of Cd2+ concentration must be considered during any assay. It has been observed that in Arabidopsis suspension cultures, treatment with 50 µm Cd2+ does not result in any differences versus untreated cultures. In the same system, however, approximately 80% of cells are dead at 7 and 4 d after 100 and 150 µm Cd2+ treatments, respectively (De Michele et al. 2009). In tobacco cell cultures and onion root apical cells, high doses of Cd2+ induce a necrotic type of cell death, while 50 µm Cd2+ treatment results in characteristics typical of PCD (Fojtová & Kovarík 2000; Behboodi & Samadi 2004). As shown in Supporting Information Fig. S2a, plant growth was greatly inhibited at Cd2+ concentrations higher than 50 µm, and some seeds could not even germinate when 200 µm Cd2+ was added to the MS agar plates. FDA and Hoechst 33258 experiments also indicated that 200 µm Cd2+ caused high toxicity in 14-day-old Arabidopsis seedlings, which are suitable for investigating the toxic effects of Cd2+. Similarly, 100 µm Cd2+ was adequate for observing the process and investigating the mechanism of Cd2+-induced PCD (Supporting Information Fig. S2b–d).

Recently, Palmieri et al. (2010) identified possible targets for S-nitrosylation in the mitochondria of Arabidopsis. This supports a model of cross-talk between NO/ROS and mitochondria in the activation of stress-related responses in plants. In the current study, we found that the induction of NO occurred first in the mitochondrial regions of protoplasts and subsequently in chloroplasts and even the whole cell as treatment time increased (Supporting Information Fig. S3). Mitochondria have been thought to be the primary source of NO in dark-induced senescence and plant stress-related responses (Guo & Crawford 2005; Planchet et al. 2005). Jasid et al. (2006) found that chloroplasts are also a source of NO in plants. Because DAF-FM DA is a membrane-permeable fluorogenic reagent, there are two possibilities about the appearance of NO generation in chloroplasts and other organelles. Firstly, it may be due to NO and DAF-FM DA diffusion from the mitochondria. Secondly, other organelles, such as chloroplasts, may participate in Cd2+-induced NO production as the treatment time increased. The subcellular localization of NO accumulation requires additional research in the future.

We have previously reported that changes in the distribution and mobility of mitochondria serve important functions in Cd2+-induced cell death (Bi et al. 2009). We have also found that pretreatment with the MPTP inhibitor CsA depresses the mitochondrial dysfunction and caspase-3-like activation induced by Al stress (Li & Xing 2011). In order to investigate the relationship between NO and caspase-3-like protease in Cd2+-induced PCD, we measured Cd2+-induced caspase-3-like activity after pretreatment with cPTIO. In the presence of cPTIO, Cd2+-induced caspase-3-like activity was reduced significantly (Fig. 4). This result is in agreement with a previous report that in Papaver, ROS/NO scavengers alleviated SI-induced activation of DEVDase/caspase-3-like activity (Wilkins et al. 2011). Our results further showed that adding cPTIO could promote plant growth and cell viability, reduce the amount of cells undergoing PCD, and prohibit the Cd2+-induced cell death (Fig. 5).

NO promotes caspase-3-like activation via MPK6-mediation

NO-dependent activation of caspase-3 has been widely reported in animal cell apoptosis (Sengupta et al. 2009; Adamczyk et al. 2010). The mechanism of NO-medicated caspase activation, however, is not yet clear. It has been reported that a MAPK cascade is involved in Cd-induced cell death (Xu et al. 2011) and that NO is also necessary for activation of the MAPKs (Pagnussat et al. 2004; Zhang et al. 2007; Besson-Bard et al. 2008). In this study, we have investigated whether NO might act as a signal for Cd2+-induced MAPKs activation. Plant MAPKs have high homology to mammalian ERK1/2 MAPKs. Antiphospho-p44/42 antibody is an established technique to measure the phosphorylated active MAPK forms in Arabidopsis (Wang et al. 2010). The study utilized the same antibody. Following pretreatment with the NO scavenger cPTIO, the Cd2+-induced activation of MAPKs was greatly suppressed (Fig. 6). In agreement with these results, Liu et al. (2010) have reported that Cd2+ activates Arabidopsis MPK3 and MPK6 via the accumulation of ROS, and De Michele et al. (2009) found that in Arabidopsis suspension cultures, Cd2+-induced H2O2 production occurred after NO production because Cd2+-induced NO production negatively affected the CAT and ascorbate peroxidase (APX) capacity. It has also been reported that NO promoted ROS accumulation by modulating the activity of NOX and antioxidative enzymes (Xu et al. 2010). Based on these findings, it is reasonable to conclude that NO is involved in the Cd2+-induced activation of Arabidopsis MAPKs and it is likely that NO and ROS somewhat act synergistically to regulate the MAPK signalling pathway.

As our data suggested that Cd2+-induced NO regulated caspase-3-like and MAPK pathway activation (Figs 4 & 6), we confirmed that MAPK activity was involved in triggering Cd2+-induced caspase-3-like activity in Arabidopsis. In the presence of the common MAPK inhibitor PD98059, Cd2+-induced caspase-3-like activity was greatly decreased (Fig. 7a). Furthermore, as shown in the mpk-6 deletion mutants in Fig. 7b,c caspase-3-like activity was suppressed during Cd2+ treatment. Although both 44 and 47 kDa MAPKs were activated by Cd2+ stress, these two proteins could have different functions in Cd2+-induced PCD. This is consistent with the previous observation that MPK3 and MPK6 play different roles in basal and oligogalacturonide or flagellin-induced resistance against Botrytis cinerea (Galletti, Ferrari & De Lorenzo 2011). All these results are similar to previous studies that have suggested that the MAPK pathway is involved in caspase-3-dependent apoptosis in animal cells (Kishi et al. 2010) and plants (Li, Šamaj & Franklin-Tong 2007; Bosch et al. 2008). Finally, we observed that AtMPK6 deficiency and pretreatment of PD98059 partially suppressed Cd2+-induced cell death (Fig. 8). This confirmed that MPK6 is involved in Cd2+-induced PCD.

In conclusion, NO is a key signalling molecule in Cd2+-induced PCD in Arabidopsis. Our results suggested that a caspase-3-like protease activation was associated with Cd2+-induced Arabidopsis PCD and that Cd2+-induced NO production contributed to this activation. MPK6 acted downstream of NO and was involved in NO-regulated caspase-3-like activation. Wang et al. (2010) proved that MPK6 modulated NO production in H2O2 stress. Consequently, the cross-talk between NO and MPK6 suggested that maybe there was an amplification loop in the NO signalling pathway during Cd2+-induced Arabidopsis PCD, which would be interesting to investigate further in the future.

ACKNOWLEDGMENTS

We thank Dr Shengjun Nie of our group for the ECFP-DEVD-EGFP transgene construction and plant screening. We appreciate the members of our group for the help in this work. This research is supported by the Program for Changjiang Scholars and Innovative Research Team in University (IRT0829), the Key Program of NSFC-Guangdong Joint Funds of China (U0931005), and the National High Technology Research and Development Program of China (863 Program) (2007AA10Z204).

Ancillary