Introduction

Apoptosis is one of the forms of programmed cell death (PCD) that induces characteristic morphological and biochemical alterations in cells and results in the orchestrated disassembly of the cell (Xu et al. 2004). Being essential for cell and tissue homeostasis and specialization (Chen and Dickman 2004), PCD also plays an important role in mediating plant adaptive responses to the environment, and was experimentally proved to occur in response to salinity, cold stress, waterlogging and hypoxia (Katsuhara and Kawasaki 1996; Pennell and Lamb 1997; Huh et al. 2002). While the regulatory mechanisms of PCD in animals are fairly well known and mostly depend on caspase protease activity (Lam et al. 2001), the apoptotic machinery and signal-transduction pathways of PCD in plants remain unclear. A key role for caspase-like proteases has been suggested (Hatsugai et al. 2004; Chichkova et al. 2004), and several reports of plant proteases with caspase-like properties that functionally mimic caspase activity in animals have been recently published (Watanabe and Lam 2004; Rojo et al. 2004).

Being upstream of the PCD signaling cascade, membrane-transport processes play a pivotal role in PCD in animal tissues (Gulbins et al. 2000; Panayiotidis et al. 2006). Thus far, no comprehensive electrophysiological study has been undertaken to investigate PCD-related membrane-transport processes in plants. However, ion fluxes are known to be the earliest detectable signaling event in plant defense responses (Clough et al. 2005). Thus, it appears that the physiological role of plant ion channels and ion transport in PCD-related processes is as important as their activities in other cellular responses to the environment (Zimmermann et al. 1999; Khurana et al. 2005).

Recent studies have suggested that transgenic expression of animal anti-apoptotic genes may significantly improve a plant’s ability to tolerate a variety of biotic and abiotic stresses (Dickman et al. 2001). Expression of animal anti-apoptotic genes suppressed the extensive plant cell death caused by pathogens, and enhanced resistance to abiotic stresses such as wounding, heat, salt, drought, cold, UV-B, and oxidative stress (Mitsuhara et al. 1999; Dickman et al. 2001; Lincoln et al. 2002; Qiao et al. 2002; Xu et al. 2004; Chen and Dickman 2004; Li and Dickman 2004). Although a direct practical application of this research is unlikely due to various social and regulatory constraints, knowledge of the mechanisms of action of these genes in plants and specifically, the role of ion flux signaling in the biological activity of apoptotic genes, may be crucial for our understanding of the role that apoptosis plays in plant stress responses.

This work is aimed at addressing the above issue. Unlike previous studies, we used transient expression of an anti-apoptotic gene CED-9 from Caenorhabditis elegans by means of a plant-virus-based vector. The benefits of virus-based transient gene expression versus transgenic plants include rapid engineering, high levels of desired protein and its systemic spread in plant tissues within a short period of time (Scholthof et al. 1996). Using a range of electrophysiological techniques, we show that expression of CED-9 anti-apoptotic gene in tobacco leaf mesophyll increased salt and oxidative stress tolerance by altering ion flux patterns across the plasma membrane. The identity of specific ion transporters was further investigated in pharmacological experiments. A relationship between the discovered pattern of ion signaling initiated by the activity of the anti-apoptotic gene and mechanisms of PCD in plants is discussed.

Materials and methods

Amplification and cloning of CED-9 gene into the plant virus-based vector

Copy DNA of the anti-apoptotic C.elegans CED-9 gene was obtained from R. Horvitz, Massachusetts Institute of Technology, Cambridge, MA, USA. PCR primers were designed based on the available CED-9 sequence to introduce the Pme1 restriction site (underlined) at both the 5′ and 3′ ends of the gene: 5′GTTTAAACAATGACACGCTGCACGGCGG 3′ (homologous primers) and 5′ GTTTAAACTTACTTCAAGCTGAACATCAT 3′ (complementary primer). The PCR-amplified CED-9 gene was directly cloned into the TOPO XL vector (Invitrogen, Carlsbad, CA, USA). XL/CED-9 plasmid was digested with PmeI and the gel-purified CED-9 fragment was subcloned into the PVX-based vector pP2C2S (obtained from D. Baulcombe, Sainsbury Laboratory, Norwich, UK), which had been linearized with EcoRV and dephosphorylated with calf intestinal alkaline phosphatase (Promega, Madison, WI, USA). The resulting plasmid was designated PVX/CED-9 (Fig. 1a). The correctness of all recombinant constructs was verified by nucleotide sequencing.

Fig. 1
figure 1

a Diagram of PVX/CED-9 vector. Closed arrow duplicated PVX CP subgenomic promoter. Open arrow PVX CP subgenomic promoter. TGB triple gene block. b Characteristic symptoms of PVX infection on wild type and PVX/CED-9 N. benthamiana leaves (8 days after inoculation). c RT-PCR products amplified from plants infected with PVX/CED-9 using primers surrounding MCS of pP2C2S. Expected size of the PCR product is ∼1.1 kb. M, Lambda DNA EcoRI/HindIII marker (Promega): 21.2, 5.1, 3.5, 2, 1.9, 1.5, 1.3, 0.9, 0.8 and 0.5 kb; 1–3, RT-PCRs from different N. benthamiana plants infected with PVX/CED-9; 3 and 4, RT-PCRs from plant infected with wild type (WT) PVX and from a non-inoculated plant, respectively. d Western blot analysis. Membranes were probed with an antibody to CED-9. M, ColorBurst Molecular weights (kDa) protein marker (Sigma): 90, 65, 40, 30, 20, 13, and 8; 1 and 2, extracts from different plants infected with PVX/CED-9; 3 and 4, extracts from plants infected with WT PVX and from non-inoculated plant, respectively

Full size image

Transcript preparation and inoculation of plants

PVX/CED-9 plasmids were linearized using restriction enzyme SpeI, and capped transcripts were generated from cDNA clones using Ambion’s T7 mMessage Machine kit (Ambion, Austin, TX, USA). The transcripts were mechanically inoculated onto fully expanded leaves of Nicotiana benthamiana. Inoculated plants were grown in a standard potting mix (Shabala et al. 2000) at 25°C with a 16 h light period in the containment greenhouse facilities at the School of Agricultural Science, University of Tasmania, Australia or Molecular Plant Pathology Laboratory, USDA/ARS, Beltsville, MD, USA.

Reverse transcription polymerase chain reaction (RT-PCR) and western blot analysis

Total nucleic acids were extracted from non-inoculated symptomatic leaves of N. benthamiana plants using TRIzol Reagent (Invitrogen, Carlsbad, CA, USA). RT and PCR were performed in a single reaction mixture using the Titan One Tube RT-PCR kit as described by the manufacturer (Roche Molecular Biochemicals, Chicago, IL, USA). The CED-9-specific primers as (above) or the following primer pair derived from the pP2C2S region surrounding multiple cloning site (MCS), were used in the amplification reaction: 5′AATCACAGTGTTGGCTTGC3′ (homologous primer) and 5′GTGGTAGTTGAGGTAGTTGAC 3′ (complementary primer). The PCR fragments were fractionated on a 0.8% agarose gel. The integrity of RT-PCR products were verified by automated nucleotide sequencing. For immmunoblot analysis, ∼100 mg of leaf tissue was homogenized in 1× PBS buffer containing protease inhibitor cocktail (Sigma Chemical Co., St Louis, MO, USA), centrifuged, mixed with 2× Laemmlli loading buffer and electrophoresed on pre-cast 10–20% Tris–Glycine gels (Invitrogen), following by blotting onto a nitrocellulose membrane (Invitrogen). Membranes were probed with polyclonal antibodies to CED-9 (Santa Cruz Biotechnology Inc., Santa Cruz, CA) and developed with the BCIP/NBT Substrate System (KPL Inc., MD, USA).

Greenhouse experiments

One week after inoculation, plants were subjected to salinity stress (daily irrigation with 200 mM NaCl). The experiment was carried out as a randomized block design, with three blocks containing six treatments × 2 plants each. Visual assessment was performed on a daily basis, and stomatal conductance, leaf chlorophyll content and maximal photochemical efficiency of PSII (chlorophyll fluorescence F v/F m ratio) were measured from both fully matured and young (recently unfolded) tobacco leaves at the end of the experiment as described in our earlier publications (Smethurst and Shabala 2003; Pang et al. 2004).

Non-invasive ion flux measurements

Net K+ and H+ fluxes were measured non-invasively using the microelectrode ion flux measuring technique (MIFE; UTas Consulting Ltd, Hobart). All details on microelectrode fabrication and calibration are given elsewhere (Shabala 2000; Shabala et al. 2006). Mature, but still growing leaves showing clear visual symptoms of infection were chosen for measurements. After leaves were excised, the leaf epidermis was gently removed, and mesophyll segments of about 4 × 5 mm size were cut and left floating (peeled side down) on a surface of measuring solution (0.1 mM CaCl2 + 0.2 mM KCl; pH 6 unbuffered) as described by Shabala et al. (2000) for about 3 h. One hour prior to measurement, mesophyll segments were immobilized in a Perspex holder and placed in a measuring chamber.

Electrode tips were positioned 50 μm above the leaf surface, tips aligned and separated by 3–5 μm. During measurements, electrodes were moved back and forth in a square-wave manner by a computerized stepper motor between two positions, 50 and 80 μm above the leaf surface, with 0.1 Hz frequency. Net ion fluxes were calculated from measured differences in the electrochemical potential between these two positions for each ion, as described earlier (Shabala et al. 1997; Newman 2001).

Electrophysiological protocols

Two different types of measurement were undertaken. Transient flux kinetics was measured in response to acute NaCl (200 mM) or H2O2 (1 mM) treatment applied to leaf mesophyll segments isolated from plants grown under normal conditions. Net K+ and H+ fluxes were measured from immobilized segments (see above) for 10 to 20 min to ensure the absence of any transient trends. A NaCl or H2O2 stock was then added to the measuring chamber, thoroughly mixed, and net ion flux kinetics measured for another 30–40 min. Steady-state net K+ and H+ fluxes were measured from either tobacco leaf mesophyll isolated from plants grown at 200 mM NaCl for 2 weeks, or from leaf segments treated with 0.2 mM H2O2 for 24 h.

Membrane potential measurements

Conventional KCl-filled Ag/AgCl microelectrodes (Cuin and Shabala 2005) with tip diameter ∼0.5 μm were used. Measurements were taken from at least five individual leaf segments for each treatment. Membrane potentials were recorded for 1.5–2 min after initial cell penetration.

Pharmacology

Prior to measurement, leaf segments were pre-treated for 1 h with a range of known channel blockers or metabolic inhibitors. These included: (1) tetraethylammonium chloride (TEA+ 20 mM); (2) gadolinium chloride (Gd3+ 50 μM) and (3) sodium orthovanadate (Na2VO4 0.5 mM). All chemicals were purchased from Sigma.

Statistical analysis

Statistical significance of mean values was determined using the standard Student’s t test at P < 0.05 level.

Results

Symptoms and stability of the recombinant virus in plants

The recombinant PVX/CED-9 virus vector was infectious and caused characteristic PVX symptoms visibly indistinguishable from wild type virus (Fig. 1b). Symptoms on leaves of inoculated N. benthamiana plants appeared 7–10 days post-inoculation and subsequently developed into a systemic infection. The engineered virus remained stable and produced CED-9 protein for at least 1 month after inoculation (monitored by Western blotting and RT-PCR).

Transient expression of C. elegans CED-9 gene in plants

RT-PCR using both CED-9 specific- and pP2C2S-derived primers confirmed the presence of target CED-9 RNA in plants, as well as the recombinant nature of the PVX vector in vivo. Amplified fragments of expected size were repeatedly obtained from different N. benthamiana plants infected with PVX/CED-9 (Fig. 1c), and the identity of RT-PCR product was verified by nucleotide sequencing. Western blots probed with CED-9-specific antibody immunologically confirmed the presence of the CED-9 protein product in PVX/CED-9 plants: the revealed protein bands corresponded to the predicted molecular mass of CED-9 (32.1 kDa). No banding pattern was observed from plants inoculated with WT PVX or from non-inoculated plants (Fig. 1d).

CED-9 expression ameliorate detrimental salinity effects

Two weeks of salt stress (200 mM NaCl) had a significant impact on phenotype, plant growth and physiological characteristics. Both non-inoculated and WT PVX-virus plants treated with NaCl were stunted and had a large number of chlorotic leaves (Fig. 2b, c), with no apparent difference between these two treatments. At the same time, plants infected with recombinant PVX/CED-9 virus were of larger size and looked healthier (Fig. 2a). The PVX/CED-9 plants had significantly higher amounts of chlorophyll compared with the NaCl-treated control and WT PVX plants (Fig. 3a). Interestingly, chlorophyll content was also significantly (P < 0.05) higher in WT PVX leaves compared with non-inoculated plants under saline conditions. The possible reason for this may be that viral infection per se caused a substantial oxidative stress in plants at the initial stages of inoculation. Thus, it is logical to expect that PVX infected plants would have better ROS scavenging (detoxification) systems, and are better adapted to the subsequent salinity stress. CED-9 plants also showed significantly higher photochemical efficiency (as evident from chlorophyll fluorescence F v/F m values; Fig. 3b), although, no significant difference in stomatal conductance (g s) values between control, WT PVX and CED-9 leaves was found (data not shown). NaCl caused about a 2.5-fold increase in stomatal resistance in all treatments.

Fig. 2
figure 2

Effect of salt stress (watering with 200 mM NaCl for 2 weeks) on plants expressing CED-9 gene, WT PVX and non-inoculated controls

Full size image
Fig. 3
figure 3

Effect of salt stress on leaf photosynthetic characteristics in control leaves, leaves infected with WT PVX virus, and PVX/CED-9 leaves. a Total chlorophyll content; b maximal photochemical efficiency of PSII (F v/F m value). Mean ± SE (n = 6 and 12, respectively). NT Non-treated with NaCl; NaCl watered with 200 mM NaCl for 2 weeks

Full size image

CED-9 effects on NaCl-induced ion flux responses

Acute salt stress (200 mM NaCl) caused an immediate and massive K+ efflux from tobacco mesophyll (Fig. 4a). No significant difference was found between control and WT PVX plants, while in PVX/CED-9 leaves, the magnitude of NaCl-induced K+ efflux was significantly less (∼40% reduction; Fig. 4a). Salinity treatment also induced a gradually increasing H+ efflux (Fig. 4b). No significant difference, however, occurred in transient NaCl-induced H+ fluxes between treatments (Fig. 4b).

Fig. 4
figure 4

Net ion fluxes measured from tobacco leaf mesophyll under saline conditions. Transient net K+ a and H+ b flux kinetics measured in response to acute salt stress; 200 mM NaCl was applied to leaves excised from plants grown in normal (non-saline) conditions. Mean ± SE (n = 6–8). Panels c and d show steady-state net ion K+ c and H+ d fluxes measured from tobacco leaf mesophyll from plants grown at 200 mM NaCl for 2 weeks. Mean ± SE (n = 30). For all ion flux measurements, the sign convention is “influx positive”

Full size image

When steady-state net K+ fluxes were measured from leaf mesophyll isolated from plants irrigated for 2 weeks with 200 mM NaCl, a significant difference was apparent between control and NaCl-treated leaves in both non-inoculated plants and WT PVX leaves. At the same time, there was no significant difference between control and NaCl-treated mesophyll segments isolated from leaves of plants transiently expressing the anti-apoptotic gene (Fig. 4c). Steady-state H+ fluxes were essentially unaffected by 2 weeks of NaCl treatment (Fig. 4d).

CED-9 effects on ROS-induced ion fluxes

Oxidative stress (1 mM H2O2 added to leaf mesophyll) induced a gradually increasing net K+ efflux (Fig. 5a). In CED-9 plants, this K+ efflux was significantly attenuated (∼50% reduction compared with control and WT PVX treatments). Incubation in 0.2 mM H2O2 for 24 h also resulted in a pronounced K+ leak from the cell. Consistent with the transient data, CED-9 plants had a better ability to retain K+ in the cell (Fig. 5c). Oxidative stress also caused a pronounced transient net H+ uptake; the magnitude of which was significantly higher in both WT PVX and CED-9 plants compared to the control (Fig. 5b). More prolonged oxidative stress (24 h in 0.2 mM H2O2) did not affect the magnitude of steady-state net H+ fluxes in control (non-inoculated) and WT PVX leaves, but resulted in a pronounced H+ efflux from CED-9 leaf mesophyll (Fig. 5d).

Fig. 5
figure 5

Net ion fluxes measured from tobacco leaf mesophyll under oxidative stress conditions. Transient net K+ a and H+ b flux kinetics measured in response to acute 1 mM H2O2 treatment. Mean ± SE (n = 5–8). c, d Show steady-state net ion K+ c and H+ d fluxes measured from tobacco leaf mesophyll after 24 h of treatment with 0.2 mM H2O2. Mean ± SE (n = 30)

Full size image

Pharmacology

Leaf pre-treatment in known ion channel blockers and metabolic inhibitors had a significant impact on both the steady-state net ion fluxes and on the magnitude of NaCl-induced flux transients (Fig. 6). Both TEA+ (a known blocker of K+-selective channels), and Gd3+ (a known blocker of non-selective cation channels; NSCC), significantly shifted steady-state K+ fluxes towards net K+ influx (Fig. 6a), with Gd3+ having much stronger effect. Both TEA+ and Gd3+ were also efficient in suppressing NaCl-induced K+ leak from leaf mesophyll (ca. 70 and 40 % inhibition, respectively; Fig. 6b). The effect of inhibitors was qualitatively similar in both non-inoculated and CED-9 plants (Fig. 6). Leaf pre-treatment in 0.5 mM vanadate (a known inhibitor of H+-ATPase), also significantly suppressed the magnitude of peak H+ flux response (data not shown), indicating a possible involvement of the plasma membrane H+-pump in mediating NaCl-induced net H+ fluxes.

Fig. 6
figure 6

Pharmacology of salt-induced (200 mM NaCl) ion flux responses from tobacco leaf mesophyll. Steady-state net K+ fluxes a and peak K+ efflux b are shown from control and PVX/CED-9 leaves after 1 h of pre-treatment in 20 mM TEACl or 50 μM GdCl3. NT Not-treated leaves (no inhibitors used). Mean ± SE (n = 5–7)

Full size image

Membrane potential changes

Consistent with previous reports (Shabala et al. 2003, 2005, 2006), NaCl-treated leaves were significantly depolarized compared with non-treated leaves. However, the magnitude of this depolarization was dramatically different between treatments (Table 1). While in the control and WT PVX plants, two weeks of irrigation with 200 mM NaCl resulted in a dramatic membrane depolarization (by 50.7 ± 3.2 and 65.2 ± 4 mV, respectively), in CED-9 leaves, the magnitude of this depolarization was only 9.8 ± 3.5 mV (Table 1). Qualitatively similar results were found for oxidative stress; 24 h of 0.2% H2O2 treatment significantly depolarized the plasma membrane of control and WT PVX mesophyll cells (by 8.2 ± 1.4 and 18.5 ± 1.3 mV, respectively), but hyperpolarized the membrane of CED-9 cells (by 6.6 ± 1.3 mV). These findings implicate the involvement of voltage-sensitive ion transporters in mediating both NaCl- and ROS-induced stress responses.

Table 1 Steady-state values of membrane potential of tobacco mesophyll cells in control and plants irrigated with 200 mM NaCl for 2 weeks
Full size table

Discussion

CED-9 anti-apoptotic genes enhance plant salt tolerance by preventing NaCl-induced K+ efflux from salinised leaf tissues

Given the central role of membrane-transport process in PCD, surprisingly little is known about specific ion transporters mediating this process and the possible effects of anti-apoptotic genes on their activity. Earlier Qiao et al. (2002) showed that NaCl treatment disrupted cell vacuole, and that expression of an anti-apoptotic Bcl-xl gene significantly delayed this process in tobacco suspension cells by some “unknown mechanism”. Both salt-sensitive mutants of yeast (cnb1Δ) and Arabidopsis (sos1) exhibited substantially more profound PCD symptoms, indicating that salt-induced PCD is mediated by ion disequilibrium (Huh et al. 2002). Yet, no electrophysiological evidence was presented.

One of the earliest detectible responses to salt stress is a massive K+ efflux, observed in both root (Shabala et al. 2003, 2005, 2006; Cuin and Shabala 2005; Chen et al. 2005) and leaf (Shabala 2000; Shabala et al. 2000, 2006) tissue. Such a large K+ efflux reduces the intracellular K+ pool (Cuin et al. 2003; Carden et al. 2003), significantly impairing cell metabolism. Consistent with the key role of K+ homeostasis in salt-tolerance mechanisms (Maathuis and Amtmann 1999), a reduction of K+ efflux correlates with increased salt-tolerance (Carden et al. 2003; Chen et al. 2005, 2007). In this work, we report that N. benthamiana plants, transiently expressing CED-9 anti-apoptotic gene from C. elegans, conferred higher salt tolerance (Figs. 2, 3). We attribute this to the ability of CED-9-expressing cells to better maintain intracellular K+ homeostasis by preventing NaCl-induced K+ leak from salinized leaf tissue (Fig. 4a). Consistent with the transient K+ flux data (Fig. 4a), CED-9 leaves showed the smallest steady-state K+ efflux in control (no salt) conditions (Fig. 4b). Assuming a cytosolic K+ concentration of around 150 mM (Cuin et al. 2003), the Nernst potential value for 0.2 mM K+ in the bath (as in this study) is E K = −167 mV. At the same time, membrane potential values in all treatments did not exceed −140 mV (Table 1), favoring passive K+ efflux under our experimental conditions. The fact that CED-9 leaves had the smallest K+ efflux (Fig. 4c), despite having the least negative membrane potential (Table 1), implies an ability of CED-9 cells to efficiently control plasma membrane K+ efflux channels, as well as a direct involvement of plasma membrane K+ efflux channels in salt stress-induced PCD in plants.

CED-9 controls both specific K+-permeable and NSCC channels in mesophyll plasma membrane

Our previous studies have suggested that NaCl-induced K+ efflux across the plasma membrane is mediated by at least two transport systems, namely outward-rectifying K+ permeable channels (KOR) and non-selective cation (NSCC) channels (Shabala et al. 2005, 2006). CED-9 appears to affect the activity of both these K+-transport systems. Pharmacological experiments suggested that steady-state net K+ flux was mediated mainly by Gd3+-sensitive NSCC channels (Fig. 6a), while NaCl-induced K+ efflux was dominated by KOR channels (as evident from TEA+ experiments; Fig. 6b), with a relatively smaller contribution from NSCC (Fig. 6b). These pharmacological results are in a good accordance with membrane potential data (Table 1). The observed ∼40% reduction in the magnitude of NaCl-induced K+ efflux in CED-9 plants (Fig. 4a) can be attributed to a much less membrane depolarization (9.8 ± 3.5 vs. 50.7 ± 3.2 mV for CED-9 and control, respectively; Table 1), implicating the involvement of voltage-sensitive KOR channels in the regulation of PCD and anti-apoptotic activity of CED-9.

CED-9 ameliorates detrimental effects of oxidative stress on plasma membrane transporters

Reactive oxygen species (ROS) are known to be primarily responsible for the impairment of cellular function under numerous abiotic and biotic stress conditions, including salinity (Zhu 2002). It has been suggested that anti-apoptotic genes suppress ROS generation in chloroplasts and mitochondria (Pennell and Lamb 1997; Xu et al. 2004). However, these conclusions are based on visual observations such as a lack of decoloration in transgenic leaves under stress conditions (e.g. Xu et al. 2004; Qiao et al. 2002). Here we report that a transiently expressed anti-apoptotic CED-9 gene increased total chlorophyll content in salt stressed leaves (Fig. 3a), resulting in substantially higher leaf photosynthetic activity (Fig. 3b). We attribute these effects to the ability of the anti-apoptotic gene to control ionic homeostasis in ROS stressed cells (Fig. 5).

Increased ROS production has been long known to detrimentally affect cellular ionic homeostasis by causing lipid peroxidation and impairing membrane integrity, making membranes more permeable to electrolytes and, specifically, to K+ (McKersie et al. 1982). At the same time, chlorophyll biosynthesis is critically dependent on the intracellular K+ level (Marschner 1995). Here we show that, as a result of H2O2 treatment, tobacco mesophyll cells had a clearly pronounced K+ efflux (Fig. 5a, c), and that CED-9 plants were capable of significantly attenuating this detrimental effect of oxidative stress in both short-term (Fig. 5a) and long-term (Fig. 5c) experiments. This can explain the higher photosynthetic activity (Fig. 3b) and chlorophyll content (Fig. 3a) in PVX/CED-9 plants compared to control or WT PVX.

Based on our data, the following tentative scenario can be offered. Under saline conditions, Na+ enters the cell through NSCC, causing (amongst other things), a significant membrane depolarization and ROS production (especially in green photosynthesizing cells). This will result in a dramatic reduction of the cytosolic K+ pool, caused by a massive K+ efflux through both depolarization-activated KOR channels (Shabala et al. 2006) and ROS-activated (Demidchik et al. 2003) NSCC channels. In animal tissues, caspase activity is significantly increased by a low cytosolic K+ content (Hughes and Cidlowski 1999). Assuming plant caspase-like proteases are regulated in a similar way, a decrease in cytosolic K+ pool may activate caspase-like proteases, leading to Programmed Cell Death. Accordingly, a plausible explanation for the ameliorating effect of animal anti-apoptotic CED-9 genes is that their expression in the leaf mesophyll may block (either directly or indirectly) the permeability of NSCC, reducing Na+ uptake into the cell and preventing NaCl-induced depolarization and the resulting K+ leak through KOR. In addition, CED-9 blockage of the KOR is also possible. Altogether, this maintains the optimal intracellular K+ homeostasis and prevents activation of caspase-like proteases (and, hence, PCD).

In summary, this work represents the first attempt to link ion flux “signatures” and apoptosis in plants. We have shown that expression of animal anti-apoptotic CED-9 gene in tobacco leaf mesophyll caused a significant alteration in cell electrophysiological characteristics and, specifically, activity of plasma membrane K+ and H+ transporters. The specific nature of these transporters should be further investigated in patch-clamp experiments and by expressing anti-apoptotic genes in Arabidopsis knock-out mutants.