Introduction

Cadmium (Cd) is one of the most toxic heavy metals for herbaceous (DalCorso and others 2008) and woody plants (Elobeid and others 2012; Polle and others 2013). It exerts adverse effects on various physiological processes such as photosynthesis, respiration, nitrogen metabolism, and nutrient uptake, leading to growth retardation and even plant death (di Toppi and Gabbrielli 1999; DalCorso and others 2008). Cd2+ is suggested to enter plant cells through non-selective cation channels and Fe2+, Ca2+, and Zn2+ transporters/channels such as iron-regulated transporter (IRT) 1, zinc-regulated transporter/IRT-like protein, and natural resistance-associated macrophage protein family transporter (Clemens 2006; Lux and others 2011; Zhu and others 2012; Sasaki and others 2012). Excessive Cd2+ in the cytoplasm usually promotes a burst of reactive oxygen species (ROS), for example, superoxide (O ·−2 ) and hydrogen peroxide (H2O2) (Gallego and others 2012; Chmielowska-Bak and others 2014). ROS accumulation can lead to oxidative stress within the cells, including harmfully changing protein structures, destroying phospholipids, and eventually causing membrane damage and enzyme inactivation (Gallego and others 2012). In contrast, initially produced ROS also can be signaling molecules that regulate a large network in the cellular response to cadmium toxicity (Sandalio and others 2009; Rodríguez-Serrano and others 2009). We previously found that H2O2 activates plasma membrane (PM) Ca2+ channels, thus enhancing entry of Cd2+ into the cytosol (Sun and others 2013).

Abscisic acid (ABA) is a crucial phytohormone that regulates plant growth and development, including seed dormancy, stomatal movement, and lateral root formation (Finkelstein and others 2002). It also plays important roles in plant adaptations to various environmental stresses, including drought, salinity, and freezing (Zhu 2002; Thompson and others 2007; Guo and others 2012; Yang and others 2014). Recently, increasing numbers of reports have confirmed the physiological roles of ABA in heavy metal tolerance in herbaceous plants. For example, exogenous ABA significantly enhances cadmium tolerance by reducing transpiration rates in rice seedlings (Hsu and Kao 2003, 2005). In Atractylodes macrocephala, the application of ABA increases antioxidant enzyme activity and decreases Pb2+ content in shoot and root tissues, alleviating Pb2+-induced oxidative damage (Wang and others 2013a). Moreover, pretreatment with ABA markedly attenuates the inhibitory effect of Cd2+ on adventitious rooting in mung bean seedlings (Li and others 2014). However, the underlying mechanisms responsible for the altered Cd2+ toxicity remain largely unknown in woody plants.

As a salt-tolerant tree, Populus euphratica is usually used as a model species to explore mechanisms involved in plant responses to salinity (Chen and others 2001, 2002a, 2002b, 2003; Sun and others 2009, 2010a; Chen and Polle 2010; Han and others 2013; Ma and others 2013; Chen and others 2014; Polle and Chen 2015). However, P. euphratica is sensitive to Cd2+ stress, mainly because of a failure to activate early protective responses upon Cd2+ exposure (Polle and others 2013). Therefore, P. euphratica is considered ideal for investigating the effects of exogenous chemicals on Cd2+ detoxification in woody plants (Sun and others 2013). The present work aims to examine the possible role of exogenous ABA in the alleviation of cellular Cd2+ toxicity in P. euphratica. In this study, the effects of ABA on antioxidant enzyme activity, H2O2 accumulation, and Cd2+ uptake were investigated in Cd2+-stressed P. euphratica cells. In addition, we used a non-invasive ion flux technique to measure the cellular fluxes of Cd2+ in P. euphratica. The main objective was to examine ABA-induced alternations in ion fluxes in this poplar under cadmium stress.

Materials and Methods

Plant Material and Treatments

P. euphratica callus cells were induced from shoots as described previously (Sun and others 2010a, 2010b, 2013). The cells were grown on a Murashige and Skoog (MS) solid medium (pH 5.7), supplemented with 2.5 % sucrose, 0.25 mg L−1 6-BA (6-benzylaminopurine), and 0.50 mg L−1 NAA (1-Naphthaleneacetic acid). Cultures were maintained at 25 °C in the dark and sub-cultured every 3 weeks. Dose tests of CdCl2 and ABA effects on cell growth were examined in this study. P. euphratica cells were treated with different concentrations of CdCl2 (0, 25, 50, 100, and 200 μM) supplemented with or without ABA (5 μM; Note: ABA was able to reduce the effects of Cd2+ on viability and membrane permeability (MP) at 5 μM, Table 1). Cell growth was reduced by increasing Cd2+ concentrations in the medium after 3 weeks of culture (Supplementary Fig. S1). ABA could alleviate Cd2+ inhibition of cell growth and the ABA effect was more pronounced when the medium was supplemented with 100 μM CdCl2, as compared to the low (25 and 50 μM) or high doses (200 μM) of CdCl2 (Supplementary Fig. S1). Thus 100 μM CdCl2 was used for the following experiments in this study.

Table 1 Effect of exogenous ABA on cell viability and membrane permeability (MP) in P. euphratica cells under Cd2+ stress
Full size table

Dose tests of ABA on cell viability and MP were examined in Cd2+-stressed cells. After 15 days of transformation onto a fresh solid MS medium, cell cultures were incubated in a liquid MS medium (LMS) for a 6-h equilibration and then treated with or without CdCl2 (100 μM in LMS) in the absence or presence of ABA (0.5, 5, and 20 μM). Control cells were treated without the addition of CdCl2 or ABA. Cell samples were harvested at 72 h to examine cell viability and MP.

To examine the time course of H2O2, antioxidant enzymes (SOD, CAT, and APX) and Cd2+ fluxes, P. euphratica cultures were exposed to CdCl2 (0 or 100 μM in LMS) supplemented with or without ABA (5 μM). Cells were sampled at 0, 24, 48, and 72 h, and used to measure H2O2, antioxidant enzymes activity, and Cd2+ fluxes.

Determination of Cell Viability

Cell viability was determined with the fluorescent dye fluorescein diacetate as described by Sun and others (2010b, 2012a, 2012b). Briefly, cell samples were harvested and stained with 20 μg mL−1 fluorescein diacetate (Sigma-Aldrich, St. Louis, MO, USA) for 5 min in the dark at room temperature. Then the fluorescence of living cells was visualized under a Leica SP5 confocal microscope (Leica Microsystems GmbH, Wetzlar, Germany) with excitation at 488 nm and emission at 515–530 nm. Cell viability was examined by measuring five randomly selected fields on each slide, and for each field at least 100 cells were analyzed.

Membrane Permeability Measurement

MP was determined in terms of relative conductivity according to Wang and others (2007, 2013b) and Sun and others (2010b). The relative change in the conductivity is due to the release of soluble solutes from the disrupted membranes. In brief, callus cells (0.2 g) were cultured in redistilled water at 25 °C for 2 h, and then the conductivity (C1) was measured. Thereafter, cells were heated at 95 °C for 1 h to determine the total conductivity (C2). The electrical conductivities, C1 and C2, were measured with an electrical conductivity meter (DDSJ-318, LeiCi Co., Shanghai, China).

The MP was calculated as follows:

$$ {\text{MP }}\left( \% \right) \, = {\text{ C1}}/{\text{C2}} \times 100 $$

Enzyme Activity Assays

Callus samples (0.2 g) were ground to a fine powder and homogenized with 2 mL of 50 mM potassium phosphate buffer (pH 7.0) containing 1 mM EDTA and 1 % polyvinylpyrrolidone. The homogenate was centrifuged at 13,000 rpm for 20 min at 4 °C, and the supernatant was used for the enzyme assay. Protein concentration was determined with the Pierce BCA Protein Assay Kit (Thermo, USA). In the case of APX measurement, 1 mM ascorbic acid was added to the extraction buffer. The total activities of CAT and APX were determined as described previously (Sun and others 2010b, 2013). The activity of SOD was measured using an SOD assay kit A001-3 (Nanjing Jiancheng Bioengineering Institute, China) according to the manufacturer’s instructions. The results of this enzymatic assay were given in units of SOD activity per milligram of protein (U mg−1 protein) (Wang and others 2007, 2008), where 1 U of SOD was defined as the amount of enzyme producing 50 % inhibition of a colorimetric reaction between superoxide anion and a water-soluble tetrazolium salt, WST-1 (2-(4-iodophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfo-phenyl)-2H-tetrazolium, monosodium salt).

Detection of Cellular H2O2

H2O2 content was detected using the H2O2-sensitive fluorescent probe H2DCF-DA (2′,7′-Dichlorodihydrofluorescein diacetate, Molecular Probes, Eugene, OR, USA). Callus cells were incubated in 50 μM H2DCF-DA for 5 min in the dark. Then the cells were fixed on poly-l-lysine-pretreated cover slips and washed 3–4 times with LMS solution. The specific fluorescence of H2O2 produced in cells was examined with a Leica SP5 confocal microscope (Leica Microsystems GmbH, Wetzlar, Germany) with excitation at 488 nm and emission at 500–530 nm. Three-dimensional reconstructed images (maximum) of cells were used to calculate the relative fluorescence intensity using Image-Pro Plus version 6.0 software.

Visualization of Intracellular Cd2+ Levels

Cell cultures were exposed to 0 or 100 μM CdCl2 for 72 h in the absence or presence of 5 μM ABA. A Cd-specific fluorescent dye, LeadmiumTM Green AM, was used to detect Cd2+ within cells (Sun and others 2013; Han and others 2014). The stock solution of LeadmiumTM Green AM was prepared by adding 50 μL dimethyl sulfoxide to the dye. Then the stock solution was diluted 1:20 with 0.85 % NaCl. Cells were stained with LeadmiumTM Green AM for 1 h in the dark and then washed three times with 0.85 % NaCl. The fluorescence of cells was visualized under a Leica SP5 confocal microscope with excitation at 488 nm and emission at 505–530 nm. The relative fluorescence intensity was calculated with Image-Pro Plus version 6.0 software.

Measurement of Cd2+ Content

P. euphratica cells were treated with 100 μM CdCl2 for 3 weeks in the absence or presence of 5 μM ABA. Then callus cells were sampled and dried at 60 °C for 48 h. Dry samples (0.1 g) were digested with HNO3/HClO4 (85/15, v/v), and Cd2+ concentration was measured using an inductively coupled plasma optical emission spectrometer (OPTIMA 2000; PerkinElmer, USA).

Cd2+ Flux Recordings

Net Cd2+ flux was measured non-invasively using the non-invasive micro-test technique (NMT; NMT-YG-100, Younger USA LLC, Amherst, MA01002, USA). The protocols for the preparation of Cd2+-selective electrodes were followed as described by Sun and others (2013) and Han and others (2014) with modifications. Briefly, glass micropipettes (XYPG120-2; Xuyue Sci. and Tech. Co., Ltd.) with an external tip diameter of 2–4 μm were pre-pulled and silanized with tributylchlorosilane. The micropipettes were filled with a backfilling solution (10 mmol/L Cd(NO3)2 + 0.1 mmol/L KCl) and then front filled with a commercially available ion-selective cocktail (Cadmium Ionophore I, 20909, Sigma-Aldrich, St. Louis, MO, USA). An Ag/AgCl wire electrode holder (XYEH01-1; Xuyue Sci. and Tech. Co., Ltd.) was inserted into the back of the electrode to produce an electrical contact with the electrolyte solution. DRIREF-2 (World Precision Instruments) was used as the reference electrode (CMC-4). The microelectrodes were calibrated in 0, 5, 10, and 50 μM CdCl2 solution prior to the net Cd2+ flux measurements. Only electrodes with Nernstian slopes >25 mV/decade were used. The ion flux was calculated using Fick’s law of diffusion:

$$ J = - D\left( {dc/dx} \right) $$

where J represents the ion flux in the x direction, dc/dx is the ion concentration gradient, and D is the ion diffusion constant in a particular medium. The flux data were acquired with the ASET software, which is part of the NMT system, and calculated using MageFlux developed by the XuYue company (http://xuyue.net/mageflux).

Steady-State Cd2+ Flux Recording

P. euphratica cells were exposed to 0 or 100 μM CdCl2 for 0, 24, 48, and 72 h in the absence or presence of 5 μM ABA. Prior to flux measurement, cells were immobilized in the measuring solution (0 or 100 μM CdCl2, 50 μM CaCl2, pH 5.5) and equilibrated for 10 min. Then the steady-state Cd2+ flux was recorded for 10 min in each cell.

In addition, the effect of ABA pretreatment on Cd2+ flux was examined in this study. P. euphratica cells were pretreated with or without 5 μM ABA for 12 h, and then exposed to 0 or 100 μM CdCl2 for 72 h. Then the steady-state Cd2+ flux was measured with NMT.

Transient Cd2+ Flux Recording

Cells were pretreated with 5 μM ABA, 3 mM H2O2, or 5 μM ABA plus 3 mM H2O2 for 6 h, and then immobilized in the measuring solution (50 μM CaCl2 supplemented with or without 5 mM LaCl3, pH 5.5). Cd2+ fluxes were continuously recorded for 10 min prior to the Cd2+ treatment. A stock solution of CdCl2 (100 mM) was slowly added to the measuring solution until the final concentration in the solution reached 100 μM, followed by continuous recording of the Cd2+ flux for 30 min. The data measured during the first 2–3 min were discarded because of the diffusion effects of the stock addition.

Data Analysis

Statistical analyses were performed using SPSS version 17.0 software. Unless otherwise stated, differences were considered statistically significant when P < 0.05. Non-parametric statistics were used to analyze the percentage data of cell viability and MP between two independent groups (Mann–Whitney U test) or among multiple groups (Kruskal–Wallis test).

Results

ABA Alleviated Cd2+ Toxicity in P. euphratica Cells

In this study, cell viability and MP (an indicator of lipid peroxidation) were examined to quantify the Cd2+-induced toxicity in P. euphratica cells. CdCl2 (100 μM, 72 h) caused a significant decrease in cell viability but markedly increased MP (Table 1). This indicates that Cd2+ treatment resulted in a distorted and disrupted membrane, leading to an increased release of intracellular solutes and thus decreased cell viability (Wang and others 2007, 2013b; Sun and others 2010b). Of note, exogenously applied ABA reduced the effects of Cd2+ on viability (by 10–29 %) and MP (by 15–82 %), and the effects were more pronounced at 5 μM compared to the low (0.5 μM) or high dose (20 μM) (Table 1). Treatments with ABA (0.5–20 μM) had no obvious effect on either viability or MP in the absence of Cd2+ (Table 1). Furthermore, our results showed that ABA could maintain cell growth under Cd2+ stress. As shown in Fig. 1, upon 3 weeks of 100 μM CdCl2 exposure, P. euphratica cells treated with ABA (5 μM) exhibited 83 % higher fresh weight than non-ABA-treated cells. Collectively, these results indicated that ABA could alleviate Cd2+ toxicity in P. euphratica cells.

Fig. 1
figure 1

Effect of exogenous ABA on cell growth of P. euphratica under Cd2+ stress. Cells were treated with or without CdCl2 (100 μM) for 3 weeks in the presence or absence of 5 μM ABA. a Representative images showing cell performance. b Fresh weight of cells. Each column is the mean of four to five biologically independent samples, and bars represent the standard error of the mean. Columns labeled with different letters (a, b, c) indicate significantly differences at P < 0.05. Scale bar 1 cm

Full size image

ABA Increased Antioxidant Enzyme Activity and Reduced H2O2 Accumulation under Cd2+ Stress

To examine the effect of exogenous ABA on antioxidant enzymes, we measured CAT, SOD, and APX activities in P. euphratica cells after 0, 24, 48, and 72 h of Cd2+ stress. There was no significant difference in antioxidant activities when ABA and Cd2+ treatments were initiated (0 h, Fig. 2a–c). CdCl2 (100 μM) stress accelerated the activities of the three enzymes at 24, 48, and 72 h (Fig. 2a–c). It was notable that ABA increased the activity of CAT and APX in Cd2+-stressed P. euphratica cells at the three time points (24, 48, and 72 h) (Fig. 2a, c). ABA + Cd2+ treatment resulted in significantly higher SOD activity than Cd2+ treatment at 72 h, although their activities were similar at 24 and 48 h (Fig. 2b). Compared with control cells, ABA alone had no significant effect on the activities of the three antioxidant enzymes in the absence of Cd2+ stress (Fig. 2a–c).

Fig. 2
figure 2

Effect of ABA on activity of antioxidant enzymes in P. euphratica cells under Cd2+ stress. a CAT; b SOD; c APX. Cells were subjected to 100 μM CdCl2 for 0, 24, 48, and 72 h in the presence or absence of 5 μM ABA. Each column is the mean of three to four biologically independent samples, and bars represent the standard error of the mean. Columns labeled with different letters (a, b, c, d, e) denote significantly differences at P < 0.05

Full size image

In addition, a specific H2O2 probe, H2DCF-DA, was used to detect H2O2 accumulation in P. euphratica cells after 0, 24, 48, and 72 h of Cd2+ stress. P. euphratica cells exhibited very low H2O2 level at the initiation of ABA and Cd2+ treatments (0 h, Fig. 3). CdCl2 stress resulted in a marked elevation of H2O2 and the H2O2 increased with the period of Cd2+ exposure (24–72 h, Fig. 3). Of note, ABA application (5 μM) markedly decreased Cd2+-induced H2O2 production over the observation time, 24, 48, and 72 h (Fig. 3). These results suggested that ABA could enhance the activity of antioxidant enzymes and decrease the level of H2O2 in Cd2+-stressed P. euphratica cells.

Fig. 3
figure 3

Effect of exogenous ABA on H2O2 accumulation in P. euphratica cells under Cd2+ stress. Cells were subjected to 100 μM CdCl2 with or without 5 μM ABA for 0, 24, 48, and 72 h and then stained with the H2O2-sensitive fluorescent probe H2DCF-DA. Representative images show H2O2 production (green fluorescence) in P. euphratica cells. Each value (±standard error) is the mean of three to four biologically independent experiments, and 80–100 individual cells were quantified for each treatment. The mean fluorescence values labeled with different letters are significant differences at P < 0.05. Scale bar 10 μm

Full size image

ABA Reduced Cd2+ Accumulation in P. euphratica Cells Under Cd2+ Stress

In the present study, a Cd2+-sensitive fluorescent probe, LeadmiumTM Green, was used to monitor Cd2+ accumulation in P. euphratica cells after 72 h of Cd2+ treatment. CdCl2 stress (100 μM) caused evident Cd2+-specific fluorescence in P. euphratica cells, whereas Cd2+-specific fluorescence was nearly undetectable in control cells (Fig. 4). Of note, under Cd2+ stress, the ABA-treated P. euphratica cells exhibited 33 % less fluorescence intensity compared with the cells treated without ABA (Fig. 4).

Fig. 4
figure 4

Effect of exogenous ABA on Cd2+ levels within P. euphratica cells under Cd2+ stress. P. euphratica cells were treated with 100 μM CdCl2 for 72 h in the presence or absence of ABA (5 μM) and then incubated with a Cd2+-fluorescent probe (LeadmiumTM Green AM) for 1 h. Each value represents the mean of at least 50 individual cells quantified from three to four biologically independent experiments. The mean fluorescence values labeled with different letters, a, b, and c, are significant differences at P < 0.05. Scale bar 10 μm

Full size image

The Cd2+ content was also determined in P. euphratica cells after long-term exposure to 100 μM CdCl2 (3 weeks). The Cd2+ concentration in the ABA-treated cells was 16 % less than in cells treated without ABA (Fig. 5). This result was in accord with the measurements with the LeadmiumTM Green probe, suggesting that exogenous ABA application could decrease Cd2+ accumulation within P. euphratica cells under Cd2+ stress.

Fig. 5
figure 5

Effect of exogenous ABA on Cd2+ content in P. euphratica cells under Cd2+ stress. Cells were subjected to 100 μM CdCl2 in the presence or absence of 5 μM ABA for 3 weeks. Each column represents the mean of three experiments, and bars indicate the standard error of the mean. Different letters (a, b, c) denote significantly differences at P < 0.05

Full size image

ABA Reduced Cd2+ Influx in Cd2+-Stressed P. euphratica Cells

To determine whether the reduced accumulation of Cd2+ was the result of a reduction in Cd2+ uptake in P. euphratica cells, we measured the net Cd2+ fluxes with NMT. Cd2+ flux in P. euphratica cells was extremely low or under detection limit before the addition of CdCl2 (100 μM) (0 h; Fig. 6). After exposure to Cd2+ for 24–72 h, the steady Cd2+ influx was remarkably enhanced (Fig. 6). It was notable that the Cd2+ influx increased with the period of Cd2+ exposure, reaching 33.7 pmol cm−2 s−1 at 72 h (Fig. 6). However, application of ABA (5 μM) reduced the Cd2+ influx by 56–79 % in CdCl2 (100 μM)-treated cells over the observation periods (24–72 h; Fig. 6).

Fig. 6
figure 6

Effect of exogenous ABA on steady-state Cd2+ fluxes in P. euphratica cells under Cd2+ stress. Cells were treated with 100 μM CdCl2 for 0, 24, 48, and 72 h in the presence or absence of 5 μM ABA. Steady-state Cd2+ fluxes were continuously recorded for 10 min for each cell. Each column represents the mean of 15 individual cells quantified from three to four biologically independent samples. Bars indicate the standard error of the mean. Different letters (a, b, c, d, e) denote significantly differences at P < 0.05

Full size image

The effect of ABA pretreatment on steady-state Cd2+ flux was examined in this study. P. euphratica cells were pretreated with or without 5 μM ABA for 12 h, and then exposed to 0 or 100 μM CdCl2 for another 72 h. ABA pretreatment could significantly inhibit Cd2+ uptake in CdCl2-stressed cells (Supplementary Fig. S2). This is similar to the finding when ABA and Cd2+ were applied together (Fig. 6). However, the ABA inhibition of Cd2+ influx was lower than the concomitant application of Cd2+ and ABA (Fig. 6, Supplementary Fig. S2).

The effects of ABA, H2O2, and Ca2+-channel inhibitors on transient Cd2+ flux were also investigated. P. euphratica cells were pretreated with ABA or H2O2 for 6 h and then exposed to CdCl2 shock to measure Cd2+ flux in the presence or absence of LaCl3 (an inhibitor of Ca2+-permeable channels). ABA pretreatment reduced Cd2+ influx by 86 % in CdCl2-stressed cells (Fig. 7). Of note, H2O2 pretreatment (3 mM) increased the entry of Cd2+ into P. euphratica cells (the mean value reached 35.2 pmol cm−2 s−1) (Fig. 7); however, ABA significantly reduced H2O2-elicited Cd2+ influx (Fig. 7). Additionally, in the presence of LaCl3, Cd2+ entry was markedly blocked in CdCl2-shocked cells, irrespective of H2O2 and ABA pretreatments (Fig. 7).

Fig. 7
figure 7

Effect of ABA, H2O2, and PM Ca2+ channel inhibitor (LaCl3) on transient Cd2+ fluxes in P. euphratica cells. a Transient Cd2+ kinetics. Cells were pretreated with or without ABA (5 μM) in the presence or absence of H2O2 (3 mM) for 6 h. Prior to the Cd2+ shock, a steady Cd2+ flux was continuously recorded for 10 min. Cd2+ fluxes in ABA- and/or H2O2-pretreated cells were measured in the presence or absence of LaCl3 (an inhibitor of Ca2+-permeable channels; 5 mM). Each point is the mean of six individual cells, and bars represent the standard error. The mean fluxes of Cd2+ after the addition of CdCl2 are shown in (b). Different letters (a, b, c, d) denote significantly differences at P < 0.05

Full size image

Discussion

Excessive Cd2+ can induce DNA fragmentation, chromatin condensation, and loss of cell viability in woody (Sun and others 2013) and herbaceous species (Iakimova and others 2008; Ma and others 2010), leading to programmed cell death (Iakimova and others 2008; Ma and others 2010; Sun and others 2013). In this study, cellular and subcellular ion analyses revealed that short-term (72 h) or prolonged Cd2+ exposure (3 weeks, 100 μM) resulted in evident Cd2+ accumulation within P. euphratica cells (Figs. 4, 5). This result agrees with those of a previous report (Sun and others 2013). The Cd2+ buildup, especially in the cytoplasmic region, led to a significant increase in DCF-dependent fluorescence, indicating an H2O2 burst in P. euphratica cells (Fig. 3). This Cd2+-elicited H2O2 burst has been suggested to contribute to oxidative damage (DalCorso and others 2008) and the occurrence of programmed cell death in P. euphratica cells (Sun and others 2013). In the present study, ABA reduced Cd2+ suppression of cell growth and viability in P. euphratica cells, although the alleviation of Cd2+ toxicity was dose dependent (Fig. 1, Table 1, Supplementary Fig. S1). Exogenous ABA application has been reported to reduce Cd2+ toxicity in herbaceous plant species, such as rice and mung bean seedlings (Hsu and Kao 2003, 2005; Li and others 2014). Our data showed that ABA alleviation of Cd2+ toxicity mainly resulted from the restriction of Cd2+ under Cd2+ stress (Figs. 4, 5). Similarly, Hsu and Kao (2003) showed that ABA pretreatment reduces Cd2+ content in CdCl2-stressed rice seedlings.

The NMT data indicated that the Cd2+ restriction in ABA-treated P. euphratica cells was the result of a reduced Cd2+ influx. Steady-state and transient flux recordings showed a net Cd2+ influx in Cd2+-shocked and short-term stressed P. euphratica cells (Figs. 6, 7). Moreover, the steady Cd2+ influx increased with increasing exposure time to Cd2+ stress (24–72 h; Fig. 6). ABA application significantly decreased the Cd2+ influx during the observation periods (24–72 h; Fig. 6). Similarly, ABA pretreatment could significantly inhibit Cd2+ uptake in CdCl2-stressed cells (Supplementary Fig. S2). However, the ABA inhibition of Cd2+ influx was lower than the concomitant application of Cd2+ and ABA (Fig. 6, Supplementary Fig. S2). This is likely due to the difference in the duration of ABA treatment, 12 h (ABA pretreatment) versus 72 h (Cd2++ABA). ABA may not interfere with the uptake of Cd2+ as the ABA concentration was 5 μM, which is much lower than the applied Cd2+, 100 μM.

In pharmacological experiments, the Cd2+ influx was blocked by the PM Ca2+ channel inhibitor LaCl3, indicating that the Cd2+-elicited Cd2+ influx was through the PM Ca2+ channels (Fig. 7). This finding is in agreement with Sun and others (2013), who suggested that the Cd2+ influx into the cytosol is mediated by PM Ca2+-permeable channels (Sun and others 2013). Electrophysiological evidence revealed that Cd2+ ions can be transported into cells through Ca2+ channels. Using the whole-cell patch-clamp technique, Perfus-Barbeoch and others (2002) have confirmed that Cd2+ permeates through the PM calcium channels in Arabidopsis guard cells. The permeability of Cd2+ through wheat voltage-dependent calcium channels was detected when the PM derived from root cells was incorporated into planar lipid bilayers (White 1998). Of note, H2O2 application increased the net Cd2+ influx into P. euphratica cells (Fig. 7). Moreover, the Cd2+ influx increased with the rise of endogenous H2O2 accumulation in Cd2+-stressed cells (24–72 h; Figs. 3, 6). Therefore, the Cd2+-induced H2O2 production is thought to stimulate the entry of Cd2+ into Cd2+-stressed P. euphratica cells. In accordance, Sun and others (2013) also showed that H2O2 addition enhances an immediate Cd2+ influx in P. euphratica cells.

ABA significantly reduced the influx of Cd2+ in P. euphratica cells (Fig. 6). This is presumably related to the increased activity of antioxidant enzymes, CAT, SOD, and APX (24–72 h; Figs. 2, 3). The activated antioxidant enzymes benefited P. euphratica cells to scavenge the Cd2+-elicited H2O2, and thus limited the H2O2-stimulated Cd2+ influx in a long term of stress. We hypothesize that ABA stimulated the initial generation of ROS, which up-regulated the activities of antioxidant enzymes. It has shown that ABA triggers the increased generation of ROS and activates antioxidant enzymes in water-stressed maize leaves (Jiang and Zhang 2002). Our data showed that enzyme activities of CAT, SOD, and APX increased correspondingly to the rise of H2O2 during the period of Cd2+ stress (24–72 h; Figs. 2, 3). It is likely that the Cd2+-elicited H2O2 up-regulated the activities of antioxidant enzymes. It has shown that H2O2 acted as a signaling molecule in the activation of CAT and APX in scot pine roots under Cd2+ stress (Schützendübel and others 2001). Collectively, ABA application significantly enhanced these enzyme activities, which contributed to the scavenging of H2O2 under Cd2+ stress (Figs. 2, 3). As a result, the H2O2-stimulated entry of Cd2+ was inhibited by ABA in Cd2+-stressed P. euphratica (Figs. 6, 7).

In conclusion, ABA plays a crucial role in alleviating Cd2+ toxicity in P. euphratica cells under Cd2+ stress conditions. We postulate the following model of ABA signaling in Cd2+ toxicity alleviation in this poplar (Fig. 8): In Cd2+-stressed P. euphratica, ABA up-regulates the activity of antioxidant enzymes, which scavenge the Cd2+-elicited H2O2 within cells. As a result, the entry of Cd2+ is subsequently reduced because H2O2 otherwise would stimulate Cd2+ influx through the PM Ca2+ channels in P. euphratica. The buildup of Cd2+ and Cd2+-elicited oxidative damage thus is limited in ABA-treated P. euphratica.

Fig. 8
figure 8

A schematic model of ABA alleviation of Cd2+ toxicity in P. euphratica cells. Solid arrows indicate a promotion effect, a bar indicates a repression effect, and dotted arrows indicate an unconfirmed signaling cascade

Full size image