Sodium Distribution in Salt-stressed Citrus Rootstock Seedlings

Although citrus trees are considered relatively salt-sensitive, there are consistent differences in Na and Cl tolerance among different citrus rootstocks. We grew uniform seedlings of rough lemon (RL) and the more Na-tolerant Swingle citrumelo (SC) with and without 50 mM NaCl for 42 days. Salinity reduced leaf chlorophyll and plant transpiration rate (Ep) more in RL than SC. Confocal laser scanning analyses using the Na -specific cellpermeant fluorescent probe CoroNa-Red revealed a higher capacity for Na sequestration in root tissue vacuoles of SC than in RL roots and that cell walls within the stele acted as Na traps. In leaves, however, RL had significantly higher Na-dependent fluorescence than SC. Thus, the sequestration of Na in root tissue vacuoles and its immobilization by cell walls were key contributing mechanisms enabling SC leaves to maintain lower levels of Na than RL leaves. Examination of intracellular distribution of CoroNa-Green fluorescence in SC root protoplasts verified a vacuolar localization for Na in addition to the presence of a 2to 6-mm unidentified endosomal compartment containing significantly higher Na concentrations. Although citrus trees are considered relatively salt-sensitive, there are inherent differences in Na and Cl tolerance among the many commercial citrus rootstocks (Castle et al., 2006; Maas, 1993). Salinity tolerance of citrus rootstocks is gauged by satisfactory tree growth and fruit yield under saline conditions (Castle et al., 1993) along with the amount of Cl and/or Na accumulated in foliage (Levy and Syvertsen, 2004; Maas, 1993). There are often conflicting interpretations of the relative levels of rootstock tolerance to salinity depending on variations in salinity treatments, treatment duration, scion type, growth substrate, and which type of physiological responses were measured (Levy and Syvertsen, 2004; Zekri and Parsons, 1992). Salt tolerance represents the natural ability to cope with absorbed salt ions usually involving the chelation, exclusion, or sequestration of toxic concentrations of salts away from the photosynthetic machinery in leaves (Munns and Tester, 2008). Because citrus trees are usually grafted onto rootstocks, these trees represent an opportunity to study their ability to restrict the transfer of salt ions to the shoot when used as rootstocks. Rootstock characteristics tend to be consistent even when grown as seedlings in a greenhouse (Syvertsen et al., 2010). Citrus and other fruit trees are somewhat unique in that Cl ions are considered to be more toxic than Na ions (Levy and Syvertsen, 2004). Thus, salt tolerance in citrus is usually based on Cl toxicity (Maas, 1993), but Cl toxicity is seldom studied in the absence of Na. Although the allocations of Na+ and Cl at the organ level are well described among citrus rootstocks with varying salt tolerances (Garcia-Sanchez and Syvertsen, 2006; Zekri and Parsons, 1992), the fundamental nature of the exclusion, sequestration, and/or compartmentation at the tissue and cell levels remains unresolved. Our preliminary dye fluorescence studies on Na+ and Cl distribution in citrus tissues revealed that Na dyes are more effective than Cl dyes in their fluorescence intensity and stability properties. Thus, as a first step toward understanding salt tolerance and salt ion distribution in citrus tissues, we studied Na distribution in root and leaf tissues of seedlings of citrus rootstocks with contrasting tolerances to Na ions. Despite the critical importance of active Na efflux systems for salinity tolerance in many plant species (see review by Munns and Tester, 2008), such systems have not been characterized in salt-sensitive citrus trees. Na tolerance can be achieved by reducing Na uptake by the roots, sequestration within the root tissues, decrease loading to the xylem, or by an enhanced capacity of intracellular compartmentation of Na in leaf cells (Hasegawa et al., 2000; Tester and Davenport, 2003). In addition, large amounts of ions could be excluded from the metabolic milieu by sequestration in the apoplast as reported for metal ions (Mari and Lebrun, 2006). In both roots and shoots, metabolic activities that impart salt tolerance require coordination between cell and tissue types mostly mediated by membrane-bound carriers and transporters (Cuin et al., 2011). Important mechanisms that contribute to Na tolerance are the membrane-localized Na+ transporters at the tonoplast and the plasmalemma (Tester and Davenport, 2003). Tonoplast-bound Na/ H antiporters mediate the removal of Na+ into the vacuole. Overexpression of Na/H antiporters in Arabidopsis (Shi et al., 2003), Brassica (Zhang et al., 2001), and tomato (Zhang and Blumwald, 2001) has resulted in significant improvements in salt tolerance. Using a variety of fluorescent dyes, Hamaji et al. (2009) demonstrated the almost exclusive localization of Na in the vacuolar space of root cortical cells and in root suspension cultured cells of Arabidopsis. Removing excess Na from the cytosol can also be accomplished by plasmalemmabound Na efflux transporters that extrude Na to the apoplastic space (Shi et al., 2002). The use of the apoplast as a depository of surplus or toxic ions has been demonstrated for metal ions such as nickel in roots of Leptoplax (Redjala et al., 2010). The ability of a salt-tolerant wheat cultivar, Kharchia 65 (Cuin et al., 2011), to extrude Na from the roots may contribute to its salinity tolerance. However, this strategy may be only delaying the transfer of toxic levels of Na from roots to shoots. At the whole plant level, different citrus rootstock species can differ in their allocations of Na between roots and shoots (GarciaSanchez and Syvertsen, 2006, 2009). Although the common citrus rootstock rough lemon (RL) (Citrus jambhiri Lush.) and Swingle citrumelo (SC) [C. paradisi Macf. 3 Poncirus trifoliata (L.) Raf] are both considered salt-sensitive based on Cl toxicity, SC (and other P.t. hybrid rootstocks) has the ability to limit Na accumulation in leaves compared with RL (Levy and Syvertsen, 2004). The mechanism of how such a difference in Na distribution occurs in trees grafted on these contrasting rootstocks could be related to differences in Na sequestration in their root tissues. Thus, based on molecular analyses in Arabidopsis (Shi et al., 2003; Zhu, 2000) and on determinations of intracellular Na distribution with Sodium-Green (Hamaji et al., 2009), we hypothesized a higher Na sequestration in the root vacuoles of the relatively Na-tolerant SC than in RL roots. In this report, tissue Na distribution was analyzed on a tissue dry weight basis and microscopically visualized using CoroNa-Red and CoroNa-Green dyes by laser scanning confocal microscopy. We also wanted to determine if fluorescence-based visualization of cellular Na distribution corresponded to tissue Na concentration expressed on a dry weight basis. Materials and Methods Plant material. Uniform 3-month-old seedlings of the salt-sensitive rootstock RL (Citrus jambhiri Lush) and the relatively salt-tolerant SC [C. paradisi Macf. 3 C. trifoliata (L.) Raf] were purchased from a commercial nursery and transplanted into 0.5-L plastic pots Received for publication 23 May 2012. Accepted for publication 6 Aug. 2012. To whom reprint requests should be addressed; e-mail [email protected]. 1504 HORTSCIENCE VOL. 47(10) OCTOBER 2012 filled with a peat, perlite, vermiculite (3:1:1) soilless media. Seedlings were grown in a greenhouse under natural photoperiods during the late summer when maximum photosynthetically active radiation at the plant level was 1200 mmol·m·s (LI-170; LICOR, Lincoln, NE). The average day/night temperature was 36/21 C and relative humidity varied diurnally from 40% to 100%. Plants were irrigated every other day with a dilute solution of a complete fertilizer (8N–0.7P– 6.6K) at 100 mg·L nitrogen (N) plus 6% iron chelate in a sufficient volume to leach from the bottom of all pots. Plants received 21 mg of N per week. One month after transplanting, 50 mM NaCl (salinized) was gradually added to the nutrient solution of half of the plants. To avoid an osmotic shock, salinity was increased in increments of 10 mM NaCl per day until 50 mM NaCl was achieved. The experimental design was a two rootstock 3 two salinity level (0, 50 mM NaCl) with 12 replicates per treatment. Two replicate plants were harvested and analyzed periodically during the experimental period of 42 d; there were three replicate seedlings of each rootstock that were either salinized or not for 42 d. All leaf measurements used fully expanded mature leaves from the midstem area on each seedling. As an index of leaf chlorophyll, leaf greenness was estimated using a SPAD meter (SPAD-502; Minolta Corp., Ramsey, NJ). Near the end of the experiment, whole plant transpiration (Ep) was measured gravimetrically by daily weight loss from three replicate pots sealed in a plastic bag at the base of the stem, divided by the total leaf area at harvest, and expressed in units of g·m·h. Harvested plants were separated into leaves, stems, and roots, briefly rinsed in deionized water, and oven-dried at 60 C for at least 48 h. Dried mature leaves and fibrous roots were ground to a powder and leaf N, Na, and Cl concentrations were determined in a commercial laboratory (Waters Agricultural Laboratory, Camilla, GA) and expressed as a percentage dry weight. Confocal microscopy. At periodic harvest times, water rinsed roots of RL and SC were excised 0.5 cm from the root tip and leaves were separated at the base of the petiole. Root tissue was sliced into 1to 2-mm segments and incubated in an incubation solution containing 20 mM Buffer MOPS pH 7.0, 0.5 mM CaSO4, and 200 mM sorbitol (incubation media) to recover from effects of excision and sudden changes in osmolarity and salt concentration (Davenport and Tester, 2000). Sampled leaves were sliced in segments of approximately the same size as roots at the time of sampling. For Na compartmentation analysis by fluorescent laser scanning confocal microscopy (Leica TCS SL; Leica, Heidelberg, Germany), leaf and root segments from 0 and 50 mM treated RL and SC citrus plants were incubated in 10 mM of cell permeant CoroNaRed fluorescent sodium indicator (Molecular Probes, C-24431; Invitrogen, Eugene, OR) in the incubation media above for 12 h. Tissue segments were washed three times to remove any excess fluorescent dye using the incubation solution without the CoroNa-Red. The microscope setting for detecting CoroNa-Red was lexc = 543 nm and lem over the 565to 600-nm spectral band. A series of confocal optical XY images through the thickness of the samples (total scanning volume was 150 mm with a slice thickness of 3 mm) were acquired in XYZ scanning mode using the Leica TCS SL software package. Comparison of different levels of fluorescence between cells was carried out by visualizing cells with the identical imaging settings of the confocal microscope (i.e., laser intensity, pinhole diameter, and settings of the imaging detectors). Protoplast isolation and observation. Protoplasts from SC roots were isolated by incubating root segments overnight in a cell wall hydrolytic solution containing 50 mM BTP/MES buffer (pH 5.6), 800 mM sorbitol, 2% Cellulase Onozuka RS (from Trichoderma viride; SERVA Electrophoresis GmbH, Germany), 1.25% Macerozyme R-10 from Rhizopus sp. (SERVA Electrophoresis GmbH, Germany), 0.5% pectinase from Aspergillus Fig. 2. Accumulation of Na and Cl in leaves and roots of Swingle citrumelo (SC) and rough lemon (RL) citrus rootstocks irrigated with 50 mM NaCl during 42 d. **Significant at P < 0.01. Each symbol represents one plant. Fig. 1. Visual appearance of rough lemon (RL) and Swingle citrumelo (SC) plants after 42 d with and without salt treatment. HORTSCIENCE VOL. 47(10) OCTOBER 2012 1505 niger (CALBIOCHEM, Germany), 0.1% polyvinylpyrrolidone-40, 0.05% bovine serum albumin, and 1 mM CaCl2. After 12 h of incubation, the protoplasts were washed with the hydrolytic solution without the enzymes and then incubated with 10 mM cell-permeant CoroNa-Green Sodium Indicator (Molecular Probes, C-36676; Invitrogen) for an additional 8 h. Fluorescent laser scanning images with CoroNa-Green were obtained using lexc = 488 nm and lem over the 510to 540-nm spectral band. Some of the root protoplasts were incubated with 5 mM Hoechst 33342 (Molecular Probes, H3570; Invitrogen) together with CoroNa-Green and observed under fluorescent microscopy with corresponding green and blue fluorescence filters. Photographs were taken with a Canon digital camera adapted to the microscope. Results Plant growth and ion concentration. All non-salinized seedlings appeared healthy and showed no signs of impaired growth. Salinized RL seedlings appeared chlorotic and had visible tip burn on most leaves after 40 d of salt treatment (Fig. 1). Salinized SC plants grew less than control plants but did not display any visible phytotoxic symptoms. At the end of the experimental period, root Na concentrations in SC exceeded leaf Na concentrations but root and leaf Cl concentrations were similar (Fig. 2A–B). In SC, leaf Cl concentrations far exceeded levels considered to be toxic in bearing trees (0.7%; Obreza and Morgan, 2008). In RL, both Na and Cl concentrations in salinized leaves exceeded concentration in roots, and concentrations of both ions in salinized leaves exceeded toxic levels after approximately Day 25 (Fig. 2C–D). Leaf Na in salinized RL exceeded concentrations in salinized SC leaves (Table 1); leaf Cl levels in RL were numerically higher but not significantly different from SC leaves. Although all plants were well fertilized, the largest non-salinized RL seedlings developed low leaf N below optimum (Obreza and Morgan, 2008) but all other plants had leaf N concentrations at or above optimum values of 2.7%. Salinity reduced Ep and leaf chlorophyll in both rootstock types but Ep did not differ between rootstock types. RL leaves had lower chlorophyll indices than SC leaves; however, thus salinized RL had the lowest leaf chlorophyll values. Tissue sodium compartmentation analysis. Na distribution at the tissue and cell levels in young roots and leaves was determined by CoroNa-Red fluorescence. CoroNa-Red is a water-soluble cell-permeant indicator, which fluoresces only after binding to Na. In nonsalinized control roots of both tissue types, Na-induced fluorescence was observed primarily in cortical cells scattered throughout (Figs. 3A and 3C). SC roots accumulated higher levels of Na than those of RL. The small but noticeable fluorescence signal in the absence of added Na in Figures 3A and 3C likely resulted from the inherent presence of Na in irrigation water, soils, and roots. In the salinized treatment, however, a larger number of Na accumulating cells was observed in roots of SC compared with RL roots (Figs. 3B and 3D). The higher levels of Na accumulation in SC roots were consistent throughout all samples collected and through the length of the root tips (Fig. 4). In some instances, parenchyma cells within the stele also fluoresced indicating their high Na content (arrowhead in Fig. 3B). Contrary to roots, compartmentation analysis of Na distribution in leaves from salinized trees demonstrated that RL accumulated considerably larger amounts of Na in mesophyll parenchyma cells than those of SC (Fig. 5). The large difference in Na accumulation was consistent in both lamina tissue as well as in the petiole, where accumulation occurred both in cortical cells as well as in pith parenchyma (Figs. 5B and 5D). Intracellular sodium compartmentation analysis. Fluorescent micrographs (Fig. 6A) and three-dimensional reconstruction analysis of root cortical cells (Fig. 6B–C) confirmed the vacuole as the main storage compartment for cellular Na. Figure 6B shows the three-dimensional topographic outline of several vacuoles of root cells where concave areas represent the location of organelles and cytosolic areas. A prominent feature of all cells was their fluorescent walls and the presence of a brighter cytosolic zone (arrow in Fig. 6B–C) indicating an area where Na concentration is considerably higher than in the central vacuole. The higher fluorescence intensity of this zone also was apparent in the topographic three-dimensional analysis in Figure 6C. Fig. 3. Laser scanning confocal images of roots from Swingle citrumelo (SC) and rough lemon (RL) without NaCl treatment (A, C) and after 50 mM NaCl treatment (B, D) for 42 d. Root segments were cut perpendicularly and incubated in CoroNa-Red added before visualization. Arrow in B denotes Na accumulating cell in the pericycle. Table 1. Effects of rough lemon (RL) and Swingle citrumelo (SC) rootstock and 42 d of salt treatment (0 or 50 mM NaCl) on mean (n = 3) plant transpiration (Ep), chlorophyll (SPAD) index, leaf nitrogen, sodium and chlorine concentrations of 6-month-old seedlings.