Transient apoplastic alkalinization has been discussed as a general stress factor, and is thought to represent a root-to-shoot signal that transmits information regarding an ongoing NaCl stress event from the site of the trigger to the distant plant tissue. Surprisingly, despite this importance, a number of gaps exist in our knowledge of NaCl-induced apoplastic pH alkalinization.
This study was designed in order to shed light onto the mechanisms responsible for the initiation and transiency of leaf apoplastic alkalinization under conditions of NaCl stress as supplied to roots.
An H+-sensitive fluorescence probe, in combination with ratiometric microscopy imaging, was used for in planta live recording of leaf apoplastic pH.
The use of a nonionic solute demonstrated that the alkalinization is induced in response to ionic, and not osmotic, components of NaCl stress. Tests with Cl−- or Na+-accompanying counter-ions strengthened the idea that the stress factor itself, namely Cl−, is transferred from root to shoot and elicits the pH alterations. Investigations with a plasma membrane ATPase inhibitor suggest that ATPase activity influences the course of the alkalinization by having a shaping re-acidifying effect on the alkalinization.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
Plants experience stresses, such as high salinity, mineral nutrient deficiency, drought, flooding, anoxia and cold temperatures (Roberts et al., 1984; Davies & Zhang, 1991; Jackson, 1997; Wilkinson, 1999; Berger et al., 2010; Cushman & Tester, 2010; Tester & Langridge, 2010), and have to endure biotic challenges, such as infections by pathogens or colonization by mutualistic organisms, for example, the mycorrhizal fungus Piriformospora indica (Felle et al., 2005; Schäfer et al., 2009). In order to survive this ‘environment full of stress’ (Hoson, 1998), plants must recognize, convert and finally respond to these stress factors (Augé & Duan, 1991; Davies et al., 1994; Shinozaki & Yamaguchi-Shinozaki, 1997; Felle, 2001).
The apoplast is the first plant compartment that encounters abiotic or biotic stress, making it a ‘frontier’ between the environment and the cell (Hoson, 1998; Gao et al., 2004). The apoplastic space plays a principal role in the communication of plants with the outer world and enables the plants to adapt themselves for survival within the changing environment (Davies et al., 1994; Hoson, 1998), for example by entraining their growth and development to prevailing conditions (Ehlert et al., 2011; Monshausen et al., 2011).
Transient apoplastic alkalinization has been discussed as a general stress factor (Wilkinson, 1999; Felle et al., 2005; Monshausen et al., 2011), and is thought to represent an apoplastic root-to-shoot signal (Wilkinson & Davies, 1997; Felle, 2001; Monshausen et al., 2007; Sharp & Davies, 2009) that transmits information regarding an (ongoing) stress event from the site of the trigger to the distant plant tissue (Davies et al., 1994; Felle et al., 2005; Swanson et al., 2011). Recently, a dose dependence between the intensity of NaCl stress treatment and the magnitude of the transient apoplastic alkalinization has been demonstrated by means of ratiometric in planta pH recording (Geilfus & Mühling, 2012).
Wilkinson (1999) has suggested that transient apoplastic alkalinization is of great significance to the ability of plants to survive in the field under a variety of conditions. Surprisingly, despite this importance, a number of gaps remain in our knowledge of this alkalinization (Bacon et al., 1998; Wilkinson, 1999). For example, the mechanism by which it arises is still unclear. Moreover, the alkalization that occurs is of a temporary nature, namely, the apoplastic pH peaks transiently. The mechanism responsible for this transiency is entirely unknown.
The work presented here focuses on the mechanism(s) responsible for the formation of the leaf apoplastic alkalinization that is built up in response to NaCl stress encountered at the roots. Our aim is to investigate the mechanism(s) by which the apoplastic alkalinization arises and the process contributing to the temporary nature of the alkalinization, namely, the elucidation of the reason why the apoplast is re-acidified shortly after it has been alkalized in response to NaCl stress. To address this problem, the apoplastic pH was quantified noninvasively in real time and in planta using an ion-sensitive fluorescence probe in combination with ratiometric fluorescence microscopy. The separate effect of Na+ independent of Cl− on the formation of the alkalinization was examined using counter-ions for Na+ and Cl−. The involvement of the plasma membrane H+-ATPase in the re-acidification of the apoplastic pH was studied using an ATPase inhibitor and an activator.
Materials and Methods
Cultivation of plant material
Vicia faba L., minor cv Scirocco (Saaten-Union GmbH, Isernhagen, Germany) was grown under hydroponic culture conditions in a climate chamber (14 h : 10 h day : night; 20 : 15°C; 50 : 60% humidity). Seeds were soaked in an aerated CaSO4 solution (0.5 mM) for 1 d at 25°C and subsequently placed into quartz sand moistened with CaSO4 (1 mM). After 6 d of germination, seedlings were transferred to plastic pots containing one-quarter-strength aerated nutrient solution. After 2 d of cultivation, the concentration of nutrients was increased to half-strength and, after 4 d of cultivation, to full-strength. The nutrient solution had the following composition: 0.1 mM KH2PO4, 1.0 mM K2SO4, 0.2 mM KCl, 2.0 mM Ca(NO3)2, 0.5 mM MgSO4, 60 μM Fe-EDTA, 10 μM H3BO4, 2.0 μM MnSO4, 0.5 μM ZnSO4, 0.2 μM CuSO4, 0.05 μM (NH4)6Mo7O24. The solution was changed every fourth day to avoid nutrient depletion. After 30 d of plant cultivation, in vivo pH recording was performed on growing leaves of V. faba plants.
For the purpose of ratiometric in planta measurement of apoplastic alkalinizations, a fluorescent indicator was loaded into the apoplast of intact plants as described in detail by Geilfus & Mühling (2011). Briefly, 25 μM Oregon Green 488-dextran (Invitrogen GmbH, Darmstadt, Germany) was inserted directly into the apoplast using a syringe (without a needle). By means of gentle pressure, the dye was loaded into the apoplast of the living plant according to Geilfus & Mühling (2011); any excess of loaded water originating from the dye solution exited the apoplast through the stomata. Thus, during the measurements, the apoplast was not flooded and thus not anoxic, but acted in a gaseous milieu as is normal in the apoplast of living plants. The dextran molecule conjugated to the pH-sensitive fluorophore Oregon Green 488 ensured, because of its size (10 kDa), that the dye did not access the cytosol from the apoplastic space by migration across the plasmalemma membrane, as established previously by confocal laser scanning microscopy imaging (Geilfus & Mühling, 2011, 2012).
Live imaging of apoplastic pH alterations using fluorescence microscopy
Fluorescence images were collected as a time series with a Leica inverted microscope (DMI6000B; Leica Microsystems, Wetzlar, Germany) connected to a DFC camera (DFC 360FX; Leica Microsystems) via a 20-fold magnification, 0.4 numerical aperture, dry objective (HCX PL FLUOTAR L; Leica Microsystems). An HXP lamp (HXP Short Arc Lamp; Osram, Munich, Germany) was used for illumination at excitation wavelengths of 440/20 and 495/10 nm. The exposure time was 25 ms for both channels. The dye fluorescence at both excitation channels was collected using a 535/25-nm emission band-pass filter (BP 535/25; ET535/25M; Leica Microsystems) and a dichromatic mirror (LP518; dichroit T518DCXR BS; Leica Microsystems). During the measurements, intact plants were supplied with aerated nutrient solution. Measurements were performed on dark-adapted plants in a darkened room. In order to correlate the timing of the apoplastic pH alterations with the movement of the stomatal aperture, image acquisition was modified slightly: measurements were conducted in a daylight-bright room (lamp, Professional Lighting, SON-K 400, Philips Deutschland GmbH, Hamburg, Germany; bulb, Philips SON-T Agro 400 watt, Philips Deutschland GmbH, Hamburg, Germany; light intensity was 350 μmol s−1 m−2, on average, as measured with a Li-Cor Light Meter LI-189, Lincoln, Nebraska, USA) and plants were not adapted to darkness after dye loading. However, during image acquisition, the room was darkened for < 10 s.
As a measurement of pH, the fluorescence ratio F495 : F440 of the pH-sensitive fluorophore Oregon Green 488-dextran was obtained. Oregon Green is a fluorescein derivative that has a pKa of 4.7 at which its pH sensitivity is most dynamic. Therefore, it is an appropriate dye for monitoring the acidic wall pH (Swanson et al., 2011). Image analysis was carried out using LAS AF software (version 2.3.5; Leica Microsystems). Ratio images were calculated on a pixel-by-pixel basis as F495 : F440. The background noise values were subtracted at each channel. Quantitative measurements were calculated as the ratio of the mean intensity for user-defined regions of interest (ROIs) (each ROI = 210 μm × 235 μm). For conversion of the fluorescence ratio data into apoplastic pH values, an in vivo calibration was conducted as described elsewhere (Geilfus & Mühling, 2011). In brief, 25 μM Oregon Green dye solutions were pH buffered and loaded into the leaf apoplast. The Boltzmann fit was chosen to fit sigmoidal curves to the calibration because, as explained in detail by Schulte et al. (2006), the Boltzmann equation can be derived directly from the Grynkiewicz equation (Grynkiewicz et al., 1985) describing the relationship of the analyte concentration to the fluorescence and fluorescence ratios. The fitted parameters of the Boltzmann equation include Rmin, Rmax and the apparent pK value of the calibrated indicator. Fitting was performed using Origin 7.0 (data not shown; OriginLab Corp., Northampton, MA, USA), yielding an area of best responsiveness in the range pH 3.9–6.3. When the leaves were loaded with pH buffer, all regions of the apoplast showed the same ratio signal at the same buffered pH. This uniformity suggested that the different regions of the wall did not modify the responsiveness of the dye system to pH. Despite this uniformity, the absolute pH values quoted should be viewed as approximations of the wall pH (Bibikova et al., 1998), because we cannot exclude the possibility that the buffer reaches equilibrium with the steady-state pH environment within the leaf. Nevertheless, this does not detract from the biological meaning of the way in which NaCl at the roots affects the leaf apoplastic pH (Geilfus & Mühling, 2012), because manipulation of the plasma membrane proton pump ATPase (PM-H+-ATPase) activity with fusicoccin or vanadate leads to the expected effects on the apoplastic pH (compare also Fig. 6). Thus, the Oregon Green-dextran ratio imaging approach appeared to be valid for the monitoring of physiologically relevant changes in wall pH. For pseudo-color display, the ratio was coded by hue on a spectral color scale ranging from purple (no signal) to blue (lowest detectable pH signal; pH 3.9) to pink (highest detectable pH signal; pH 6.3).
Analysis of the separate effects of Na+ and Cl− ions on the formation of the alkalinization
Hydroponic culture experiments were conducted to test the separate effects of Na+ and Cl− encountered at the roots on the formation of the transient leaf apoplastic alkalinization. The effect of Na+ independent of Cl− was examined using the macro-ion gluconate− as a sodium-accompanying counter-anion; l-cysteinium+ was used as the chloride-accompanying counter-cation (Llopis et al., 1998; Tavakkoli et al., 2011). Na-gluconate or l-cysteinium chloride was added to the nutrient solution in order to evaluate its effect on the formation of the transient leaf apoplastic alkalinization. The separate effects of Na+ and Cl− were compared with the combined effect of NaCl at the roots.
Analysis of the effects of KCl and MgCl2 on the formation of the alkalinization
The effect of Cl−-containing salts, other than sodium chloride, on the formation of leaf apoplastic alkalinization was tested using KCl and MgCl2. To separate the specific effect of Cl− from the accompanying cation (K+ or Mg2+), chloride was replaced by the macro-anion gluconate− as a counter-anion for K+ (K-gluconate) or Mg2+ (Mg-gluconate). By this means, the chloride-specific effect of the addition of KCl (or MgCl2) could be compared with the effect triggered by K-gluconate (or Mg-gluconate).
Testing for the effects of the PM-H+-ATPase on the transient shape of alkalinization
The involvement of the PM-H+-ATPase on the transient shape of alkalinization, namely the re-acidification of the pH alkalinization back to its initial level, was addressed. For this, sodium orthovanadate (Na3VO4), an inhibitor of the PM-H+-ATPase (Macara, 1980; Jia & Davies, 2007), and fusicoccin, an activator of the PM-H+-ATPase (Marré, 1979), were used. Once the transient leaf apoplastic alkalinization had been induced by the addition of NaCl to the roots, 1 mM Na3VO4 was placed onto a small area of the leaf surface; vanadate concentrations ranging from 100 μM to 10 mM have commonly been used for pretreatment in order to inhibit the PM-H+-ATPase (Shabala, 2000; Villegas et al., 2000; Yan et al., 2002; Jia & Davies, 2007). The same procedure was conducted with fusicoccin, using a final concentration of 10 μM fusicoccin; concentrations ranging from 100 nM to 20 μM fusicoccin are commonly used to activate the PM-H+-ATPase (Mengel & Schubert, 1985; Hoffmann et al., 1992; Johansson et al., 1993; Mühling et al., 1993; Fricker et al., 1997; Bibikova et al., 1998; Kosegarten et al., 1999; Lisso et al., 2011). Corresponding controls were performed using only the inhibitor (or activator) without the previous application of NaCl stress to the roots.
Determination of Na+ and Cl− concentrations in leaf dry matter
The analysis of ions was performed on 20 mg of dried plant leaves that were dissolved and boiled for 5 min in 1.6 ml of deionized water. After cooling, samples were centrifuged and the supernatant was collected. Subsequently, proteins were precipitated by washes in chloroform. Thereafter, samples were cleaned by passage through strata C-18 columns obtained from Phenomenex (Torrance, CA, USA). Thereafter, Na+ and Cl− concentrations were analyzed using ion chromatography (DX5000; Dionex, Idstein, Germany).
Confocal laser scanning microscopy
To visualize the apoplastic distribution of Oregon Green 488-dextran attached around the palisade cells, confocal laser scanning microscopy imaging via a Zeiss LSM 510 Axiovert 200M confocal laser scanning system (Carl Zeiss, Jena, Germany) was carried out. For Oregon Green excitation, the 488-nm beam line of the argon laser was chosen (emission bandwidth, 505–530 nm). Chloroplast autofluorescence was excited at 633 nm by a helium–neon laser (emission bandwidth, 655–719 nm for the Zeiss system). A Plan-Neofluar objective (Plan-Neofluar 20.0×/0.5 Ph 2; Carl Zeiss) was used for image collection. The images represent projections along the xz axes of a confocal z stack of c. 14 sections (stack size, 25.45 μm; z = 1.96 μm).
Statistics and presentation
All pH curves are continuous kinetics of the respective leaf apoplastic responses of the plant to NaCl stress. According to Felle (1994), curves representing biological replicates are not suitable for statistical treatment, that is averaging; however, all curves shown are representative recordings of six independent experiments. Mean differences between ion concentrations (Cl− or Na+) were tested. In order to maintain an experiment-wise α of P <0.05, multiple t-tests were adjusted according to Bonferroni–Holm (Horn & Vollandt, 1995). Statistical significance is indicated by lowercase letters.
NaCl-dependent apoplastic alkalinizations are transient and repetitively inducible
In response to the initiation of 70 mM NaCl stress at the root site, the leaf apoplastic pH was transiently alkalized over a period ranging from 30 to 40 min, starting c. 15–20 min after stress was initiated (Fig. 1, black curve). In order to test whether the alkalinization was multiply inducible, the NaCl stress stimulus that was added to the nutrient solution was removed by changing the complete nutrient solution. NaCl was once again added to the nutrient solution. This was repeated, each salt addition being followed by an identical alkalinization of the leaf apoplast (Fig. 1, black curve). Repeated additions of 70 mM NaCl to the roots without removal of the previous NaCl treatment primed, each time, a similar alkalinization followed by re-acidification (Fig. 1, grey curve). However, a comparison of the beginning of the formation of both pH responses (cf. removal of NaCl vs without removal of NaCl) indicated variability in the time point of the start of alkalinization. The pH changes started after 20 min (black curve) or 40 min (grey curve). A comparison of the course of both pH responses (cf. removal of NaCl vs without removal of NaCl) indicated a much shorter delay for the re-acidification under conditions in which the NaCl was removed. This was because the NaCl stress stimulus was removed at a point at which the pH had not completely returned to the initial value. However, by the end of the experiment, the wall pH had fully re-acidified.
Transient apoplastic alkalinization is not formed in response to the nonionic solutes polyethylene glycol 6000 (PEG 6000) and mannitol
In response to a decrease in the osmotic potential of the nutrient solution by the addition of the nonionic solute PEG 6000 (Lagerwerff et al., 1961) or mannitol, the leaf apoplastic pH remained stable. No transient alkalinization was formed within the apoplast (Fig. 2). This was tested with 114 g PEG 6000 l−1, representing an equivalent effect on the osmotic potential to treatment with 75 mM NaCl (according to Sümer et al., 2004) and 200 mM mannitol (Gao et al., 2004). Even the higher concentration of 228 g PEG 6000 l−1 did not trigger the formation of the alkalinization (Fig. 2).
Separate effects of Na+ and Cl− ions on the formation of the alkalinization
The effect of Na+ independent of Cl− on the formation of the alkalinization was examined using the macro-ion gluconate− as a sodium-accompanying counter-anion; l-cysteinium+ was used as a chloride-accompanying counter-cation. In response to the addition of 70 mM Na+ via Na-gluconate at the root site, the leaf apoplastic pH remained stable. No transient apoplastic alkalinization was measured within the apoplast (Fig. 3a, black curve). By contrast, the alkalinization was formed in response to the addition of 70 mM Cl− via l-cysteinium chloride at the root site (Fig. 3a, grey curve). The general effect of Cl− on the formation of the alkalinization was tested using KCl and MgCl2 as chloride-containing salts. In response to the addition of 70 mM Cl− via KCl or MgCl2, alkalinization occurred. By contrast, no transient apoplastic alkalinization arose when chloride was replaced by gluconate− using Mg- and K-gluconate as salts added to the roots (Fig. 3b).
Translocation speed of Na+ and Cl− added to the nutrient solution into the leaf
NaCl was added to the nutrient solution. After 15 min, leaf [Cl−] increased and reached a maximum 30 min after NaCl had been added to the roots (Fig. 4a). The leaf [Na+] increased significantly 45 min after NaCl had been added to the nutrient solution and remained at this concentration over the time course of the experiment (Fig. 4b).
Effect of Na+ added via foliar application on the formation of the leaf apoplastic alkalinization
The effect of Na+ independent of Cl− was examined using the macro-ion gluconate− as already described. In response to the foliar application of 10 mM or 70 mM Na+ via Na-gluconate, the pH remained stable. No transient alkalinization was primed (Fig. 5).
Involvement of the PM-H+-ATPase in shaping the apoplastic alkalinization
Transient leaf apoplastic pH alkalinizations were primed by the addition of 70 mM NaCl stress to the root site. Shortly after the pH alkalinization had begun to decrease to its initial level, 1 mM of the PM-H+-ATPase inhibitor Na3VO4 was added directly to the leaf surface, whereupon the pH decline was stopped. Moreover, the apoplastic pH was strongly alkalized (Fig. 6a, black curve). When only Na3VO4 was added to the leaf without a prior stimulation of the stress (Na3VO4 control), the apoplastic pH did not strongly alkalize in response to the inhibition of the ATPase by Na3VO4 (Fig. 6a, grey curve). In both cases, PM-H+-ATPase reacted after a delay of c. 20 min on the inactivation by the vanadate that was added to the leaf surface. The addition of 10 μM of the PM-H+-ATPase activator fusicoccin to the leaf surface immediately stopped the formation of the NaCl-induced alkalinization and the pH decreased to its initial level (Fig. 6b, black curve). When only fusicoccin was added to the leaf without a prior stimulation of the stress (fusicoccin control), the apoplastic pH decreased immediately (Fig. 6b, dark grey curve). The addition of fusicoccin to the leaf at the same time as NaCl was added to the roots resulted in an immediate decrease in the leaf apoplastic pH (Fig. 6b, light grey curve).
Stomatal aperture dynamics correlate with the time course of the alkalinization
The timing of the stomatal closure behavior was correlated with the course of the transient alkalinization. In response to the initiation of 70 mM NaCl stress at the root site, the leaf apoplastic pH was transiently alkalized (Fig. 7a, black curve). During the formation of the alkalinization, the stomata remained open. At 45 min after NaCl had been added (t = 65 min at the axis of the ordinates), stomatal closure began. By 90 min after NaCl had been added (t = 110 min at the axis of the ordinates), the stomatal pores were completely closed (Fig. 7a, grey curve; compare also with the illustrations in Fig. 7b–g).
Nature of the NaCl-induced leaf apoplastic pH increase
NaCl stress treatments of the roots of V. faba have recently been found to prime transient leaf apoplastic alkalinizations in a dose-dependent manner: the intensity of the alkalinization is dependent on the magnitude of the NaCl stress. Moreover, this systemic apoplastic alkalinization seems to be propagated from root to shoot (Geilfus & Mühling, 2012). In the current study, we have demonstrated that this apoplastic alkalinization is repetitively inducible by repeated additions of 70 mM NaCl to the roots, regardless of whether the initial NaCl stress is removed from the nutrient solution (Fig. 1, black curve) or whether it remains (Fig. 1, grey curve). This may indicate that changes in the environment in the form of an initiation of NaCl stress are followed each time by an identical reproducible systemic response being transferred to the leaves. In accordance with this, Gao et al. (2004) have shown similar repetitive pH kinetics in response to NaCl stress in the apoplast of Arabidopsis roots. However, whether the alkalinization occurs in response to an ion-specific or an osmotic-specific salt stress effect remains unclear. For this reason, we tested whether the addition of the nonionic solute PEG 6000 to the roots induced the formation of the alkalinization. Interestingly, a decrease in the osmotic potential of the nutrient solution by PEG (Lagerwerff et al., 1961) did not prime any pH alterations in the leaf apoplast within the examined period (Fig. 2). This is in disagreement with the study of Felle & Hanstein (2002), who reported that an increase in the osmolarity of the bath solution harboring petioles of leaves cut near the stem alkalized the substomatal apoplast. One possible explanation for the different results is that plants might react differently when osmotic stress is presented to the root system of a living plant (this study, Fig. 2) as opposed to conditions in which the osmolarity is changed via the petiole xylem of a detached leaf. In this study, we investigated the response that occurred at the leaf site when the stress was added at the roots. This experimental approach allowed us to measure the corresponding responses in real time and noninvasively using an intact plant system, thus taking the normal root-to-shoot communication into account. A lack of root-to-shoot communication in the approach with detached leaves might explain the differences described between the two studies. Moreover, in our design, the size of the stress inducer PEG (MW = 6000) ensured that it could not access the vascular bundle and thus did not enter the xylem or the leaf in which it might have caused unintended side effects. We further tested the nonionic solute mannitol, which can be used to induce drought stress (Munns, 2005). As with PEG 6000, no pH effects were seen with mannitol (Fig. 2). This is in agreement with the study by Gao et al. (2004), who also reported that mannitol-induced water stress did not promote a transient alkalinization in the apoplast of Arabidopsis roots, as shown using the transgenic pH indicator pHluorin. Taking these results together, the transient apoplastic alkalinization that is detectable in leaves c. 15–20 min after the initiation of stress to the root (Fig. 1) is suspected to emanate from ionic, and not osmotic, components of the NaCl stress. However, as drought has also been demonstrated to prime apoplastic pH alkalinizations (Wilkinson & Davies, 1997; Felle, 2001), Felle et al. (2005) assumed that the osmotic and salt stress inputs are each operative on a different time base. Therefore, we need to distinguish between short-term effects (minutes or, at most, hours) during which the NaCl-induced alkalinizations are detectable, and long-term effects (days; Felle et al., 2005) during which drought-related pH alterations can be observed (Hartung & Radin, 1989; Gollan et al., 1992; Bacon et al., 1998; Davies et al., 2002).
The question now arises as to whether the transient alkalinization (Fig. 1) is specifically formed in response to one of the ions, that is Na+ or Cl−, or whether V. faba does not discriminate between the single ions and perceives both Na+ and Cl− likewise with a leaf apoplastic alkalinization. To unravel the effect of NaCl on the formation of the alkalinization, Na+ or Cl− was added to the nutrient solution. The addition of 70 mM Na+ via Na-gluconate to the roots did not prime any leaf apoplastic alkalinization (Fig. 3a, black curve). However, the addition of 70 mM Cl− via l-cysteinium chloride was followed by a transient alkalinization of the leaf apoplast (Fig. 3a, grey curve). This reveals that Cl− is the environmental stress factor that elicits the leaf apoplastic alkalinization in V. faba under conditions of NaCl stress. By contrast, salt-induced alkalinizations have been reported previously to occur in the presence of the membrane-impermeable gluconate−, used instead of Cl− as the accompanying ion (Felle & Hanstein, 2002). However, in the cited study, the gluconate− was added to the cut end of petioles, representing an experimental design that might allow the gluconate− unintentionally to enter the leaf symplast via the intersected phloem. Although, in most plant species, the wounded phloem will be closed by sieve plate occlusion via forisome proteins (Gaupels et al., 2008) and callose deposition, Furch et al. (2007) have reported that, in intact V. faba plants, the cutting of the leaf tip does not induce callose deposition in the sieve elements of the main veins. Nevertheless, forisomes disperse c. 30 s after cutting the main vein and plug the sieve pores. However, it cannot be excluded that, even during this short half-minute time slot, the membrane-impermeable gluconate− may enter the phloem by crossing the injured plasma membrane of the phloem cells, where it may possibly effect the ion balance and thus the pH. As this scenario cannot be refuted, it could represent one reason for the differences described between the two studies.
In a further experiment, we tested whether a broader range of Cl−-containing salts would provide data supporting the idea that the formation of the alkalinization is a Cl−-specific response to NaCl stress. The addition of 70 mM Cl− via 70 mM KCl or 35 mM MgCl2 to the roots was followed by a transient alkalinization of the leaf apoplast (Fig. 3b). However, when chloride was replaced by gluconate− as a counter-anion for K+ (K-gluconate) or Mg2+ (Mg-gluconate), the pH effect was not initiated in the leaves (Fig. 3b). These results clearly demonstrate that Cl− at the roots is the stress factor that needs to be present to prime the alkalinization in the leaves under conditions of NaCl stress.
Timing of the formation of the transient alkalinization coincides with the delivery of detectable Cl− into the leaf
The finding that the timing of the initial formation of the alkalinization coincides with the delivery of detectable Cl− into the leaf (cf. Fig. 4a and Fig. 1) strengthens the idea proposed by Felle et al. (2005) that the stress factor itself, Cl− in this study, is transferred into the leaf and causes the pH change. This assumption is underlined by the finding that foliar application of NaCl immediately elicits a transient alkalinization (see Fig. 4 in Geilfus & Mühling, 2011), whereas NaCl treatment of the roots elicits the leaf apoplastic alkalinization with a delay of at least 15–20 min (Fig. 1). Interestingly, this is approximately the time that passes until Cl− given to the roots reaches the leaves (Fig. 4a). We assume that the driving force for the delivery of chloride from root to shoot is transpiration, whereby differences in the rate of transpiration might explain the observed variability in the time point at which alkalinization starts (cf. Fig. 1; 20 min after stress initiation, black curve, vs 50 min after stress initiation, grey curve). The finding that the foliar application of Na+ via Na-gluconate does not induce transient leaf apoplastic alkalinization (Fig. 5) further strengthens the idea that the formation of the alkalinization relies on the presence of Cl− and not on Na+ or osmotic stress at the site at which the alkalinization arises.
A conceivable explanation for the formation of the alkalinization is that chloride is taken up by a Cl−/nH+ symporter (Sanders & Hansen, 1981), resulting in an increase in the leaf apoplastic [H+] caused by the transfer of protons from the apoplast into the cytosol. Appropriately, the findings of Felle (1994) are in good agreement with the idea of a Cl−/nH+ symporter, because he demonstrated that increases in apoplastic [Cl−] cause a transient pHcyt decrease in Sinapis alba roots. Moreover, Lorenzen et al. (2004) have provided evidence that Cl− is taken up into cells after its application to the roots via NaCl. Based on these results establishing the rapid uptake of chloride, a Cl−/nH+ symporter might be the mechanism causing leaf apoplastic alkalinization under conditions of NaCl stress by the transport of protons from the apoplast into the symplast across the plasma membrane. As the apoplast-forming cells are covered by only a thin aqueous film with a low passive buffer capacity (Felle & Hanstein, 2002), such a proton translocation from the apoplast into the cytosol would be conceivable to explain the formation of the alkalinization under NaCl stress. The perception of chloride as an NaCl-related stress factor via leaf apoplastic alkalinizations might be useful for V. faba, as legume species seem to be particularly sensitive to high levels of Cl− (Marschner, 1995; Slabu et al., 2009).
Moreover, the finding that the leaf alkalinization is immediately inducible by NaCl foliar application (Geilfus & Mühling, 2011) indicates that the formation of the alkalinization does not necessarily depend on the synthesis or release of any root-sourced compounds formed in response to the addition of NaCl. Nonetheless, as evidence has been presented that root-sourced compounds affect xylem pH under NaCl stress, such root-sourced compounds might indeed play a role (Davies et al., 2002).
Cl− as a general elicitor that primes leaf apoplastic alkalinization?
The idea that Cl− primes the formation of the pH alkalinization is a case-specific explanation for the increase in the leaf apoplastic pH in response to short-term NaCl stress. As alkalinization is considered to be a general stress factor primed in response to abiotic and biotic stress events (Wilkinson & Davies, 1997; Wilkinson, 1999; Felle et al., 2005) and as a factor involved in developmental processes, such as root growth or the gravity response (Felle, 2001; Monshausen et al., 2007, 2008), the Cl−-induced formation of the apoplastic alkalinization cannot be generalized as a universal elicitor of pH signaling. By contrast, it must be seen specifically under conditions of NaCl stress. For instance, the formation of the apoplastic pH signal in barley leaves attacked by the powdery mildew fungus Blumeria graminis f.sp. hordei (Felle et al., 2004) must have a physiological basis other than a Cl−/nH+ symporter, simply because Cl− is not a stress factor during fungal infections. Another example is drought stress. A drought-induced alkalinization has been suggested to be a root-sourced signal that is carried through the xylem to the leaf (Wilkinson & Davies, 1997; Wilkinson, 1999). As summarized by Felle et al. (2005), drought has been suggested to deactivate H+ pumps and hence to cause the apoplastic pH to alkalize (Hartung et al., 1988). In addition, abscisic acid (ABA) is thought to alkalize the apoplast under conditions of drought by binding to receptors that increase cytosolic Ca2+ activity. This, in turn, activates depolarizing anion channels. Subsequently, the apoplast is thought to alkalize under conditions of drought by the export of organic acid anions, whereas uptake of the anions restores the pH (McAinsh et al., 1992; Bacon et al., 1998; Felle et al., 2000, 2005).
Transient nature of the NaCl-induced leaf apoplastic alkalinization
A transient apoplastic alkalinization interpreted as being a signal (Wilkinson, 1999; Felle et al., 2005) has been detected in various plants, such as Arabidopsis thaliana (Lorenzen et al., 2004), Hordeum vulgare (Bacon et al., 1998; Felle et al., 2004), V. faba (Felle et al., 2000; Geilfus & Mühling, 2011) and Bromus errectus (Hanstein & Felle, 1999), under various stress conditions, such as NaCl (Felle, 2001; Gao et al., 2004), drought (Bacon et al., 1998), anoxia (Felle, 2005, 2006), fungal leaf infections by B. graminis or Bipolaris sorokiniana (Felle et al., 2009) and ectomycorrhizal symbiosis (Ramos et al., 2009), and even during developmental processes, such as root growth, the gravity response or light sensing (Mühling et al., 1995; Mühling & Läuchli, 2000; Felle, 2001; Monshausen et al., 2007, 2008).
Despite this knowledge of the broad occurrence of transient apoplastic alkalinization, gaps remain in our understanding of the mechanisms by which the apoplastic alkalinization returns to its initial value after the apoplast has been alkalized in response to the (stress) stimulus. In other words, we do not know why the apoplastic pH shows a transient peak. The PM-H+-ATPase is one candidate likely to cause the re-acidification of the apoplast once the apoplast has been alkalized in response to NaCl. To address this hypothesis, we tested the influence of the PM-H+-ATPase on the transiency of the alkalinization using a PM-H+-ATPase inhibitor and activator. Incubation of a selected area of the leaf surface with the PM-H+-ATPase inhibitor vanadate (Macara, 1980) at a time at which the re-acidification had already started caused the re-acidification to stop (Fig. 6a, black curve). Moreover, the inactivation of PM-H+-ATPase yielded a continuous alkalinization up to pH 6. This suggests that the PM-H+-ATPase is responsible for the decline and thus the transient nature of the leaf apoplastic alkalinization by pumping protons into the apoplast. One idea explaining the gap between the activity of the PM-H+-ATPase and Cl− as a stress factor is that the cytosol is depolarized by the hypothesized Cl−/2H+ symporter, a process that possibly activates the PM-H+-ATPase because of the requirement of charge compensation and the need to re-establish the membrane potential. This idea is supported by the finding that positively charged ions stimulate PM-H+-ATPase activity (Sze, 1984).
The inhibition of PM-H+-ATPase activity by vanadate has the expected alkalizing effect on the wall pH; however, the wall pH might also have changed independently of the machinery that generates the transient nature of the pH change in response to NaCl. For this reason, control plants were not treated with NaCl, but only with vanadate. Inactivation of the PM-H+-ATPase had almost no effect on the leaf apoplastic pH (Fig. 6a, grey curve); however, an increase in the apoplastic pH up to pH 6, as observed in the presence of NaCl, was not seen in the absence of NaCl (cf. black and grey curves in Fig. 6a). A comparison between the two treatments implies that the NaCl-induced alkalinization would not be transient and would reach a more alkaline pH without the regulatory shaping action of the PM-H+-ATPase activity. In accordance with this finding, the activation of the PM-H+-ATPase by the activator fusicoccin (Palmgren, 2001) prevents the formation of apoplastic alkalinization (Fig. 6b, black curve), whereas the addition of fusicoccin to non-saline-treated plants has the expected acidifying effect (Fig. 6b, dark grey curve). Activation of the leaf PM-H+-ATPase at the same time as the addition of NaCl to the roots inhibits the formation of the normal alkalinization. In this case, the wall-acidifying function of the fusicoccin-stimulated PM-H+-ATPase obviously suppresses the usually observed NaCl-induced alkalization of the wall (Fig. 6b, light grey curve). An explanation for this might be the low buffering capacity within the thin apoplastic water film (Felle & Hanstein, 2002); this is apparently overcome by the fusicoccin-stimulated activity of the PM-H+-ATPase. These results indicate that the activity of PM-H+-ATPase is crucial for the transient peak in pH, which can thus be suggested to be a responsible element for the shaping of the Cl−-induced apoplastic alkalinization. This seems to be of interest, as we know from Ca2+ signaling that the shape of the peaking ion signal can provide signaling specificity (McAinsh & Pittman, 2009; White, 2009). Repeated additions of NaCl to the roots without removal of the previous NaCl treatment have been demonstrated to induce transient alkalinizations repetitively (Fig. 1, grey curve). This repetitive inducibility shows that the re-acidifying action of the PM-H+-ATPase has been overcome or is stopped after the pH has returned to the starting point, because otherwise a second (further) alkalinization would not be possible. As we have demonstrated using the mycotoxin fusicoccin that the PM-H+-ATPase can easily counteract the NaCl-induced alkalinization, the activity of the PM-H+-ATPase is unlikely to have been overcome; the PM-H+-ATPase is more likely to be regulated by some means in order to allow a second pH peak.
Timing of the apoplastic alkalinization coincides with stomatal closure
Salt stress of the roots has been shown here to elicit a transient pH change in the leaves. The timing of this change is strikingly correlated with the closure of the stomatal pores (Fig. 7). Stomatal closure represents a well-known cellular response to abiotic stress conditions (Schurr et al., 1992; Schmidt et al., 1995; Jiang & Hartung, 2008; Geiger et al., 2011) and, in particular, to the initial phase of NaCl stress (Munns & Tester, 2008). At the time at which the apoplastic pH reaches c. pH 5.25, the stomata begin to close (cf. Fig. 7a, Fig. 7b–g). This phenomenon can be discussed in the context of the interplay between the apoplastic [H+] and the apoplastic distribution of the hormone ABA. As summarized by Schroeder et al. (2001) and Jia & Davies (2007), leaf apoplastic pH can have an impact on the effect exerted by ABA on the functioning of guard cells, because increases in apoplastic pH result in a greater apoplastic accumulation of deprotonated, and thus membrane-impermeable, ABA. This accumulation can ultimately close stomata (Hartung, 1983; Hartung & Slovik, 1991; Wilkinson & Davies, 1997), because ABA may bind to an apoplastic locus of ABA perception (Hartung, 1983), inducing an internal calcium-based signal transduction cascade, causing stomatal closure (de Silva et al., 1985; McAinsh et al., 1990). However, it is also possible that the accumulated ABA reprotonates as soon as the pH re-acidifies. This would mean that the guard cells are suddenly surrounded by membrane-permeable ABAH that can enter the cytosol, in which it binds to recently identified receptors of ABA (Ma et al., 2009; Park et al., 2009), to activate a pathway that contributes to stomatal responses (Joshi-Saha et al., 2011).
However, the temporal correlation of the formation of the transient leaf apoplastic alkalinization and the closure of the stomata (Fig. 7) complements the idea proposed by Wilkinson (1999) and Felle (2001) that transient apoplastic alkalinizations represent a signal, messenger or mechanism that mediates stomatal movements under stress.
This study was designed in order to shed light onto the mechanisms responsible for the formation and transiency of the leaf apoplastic alkalinization under conditions of NaCl stress. As NaCl, but not the nonionic solute PEG 6000, triggers a transient leaf apoplastic alkalinization when added to the roots of V. faba, we conclude that the alkalinization is a response to ionic, and not osmotic, components of NaCl stress. In order to unravel the combined effect of NaCl on the formation of the alkalinization, Na+ or Cl− was added separately to the nutrient solution. This was achieved using various chloride- or sodium-accompanying counter-ions. The data presented here strengthen and extend the idea that the stress factor itself, namely Cl−, is transferred into the leaf apoplast and initiates transient alkalinization. Tests with a PM-H+-ATPase inhibitor and activator have revealed that the activity of the PM-H+-ATPase influences the course of the alkalinization once the apoplastic pH has alkalized in response to Cl−. Thus, we can conclude that the PM-H+-ATPase has a shaping effect on the alkalinization.
We thank Dr Christoph Plieth (Zentrum für Biochemie und Molekularbiologie – ZBM, University of Kiel, Kiel, Germany) for advice on fitting the calibration data to a sigmoidal Boltzmann fit. We also thank Dr Uwe Bertsch (Institute of immunology, UKSH, Kiel, Germany) for help with confocal laser scanning microscopy imaging and for providing access to the microscope. We are grateful to Stephanie thor Straten for conducting ion chromatography.