Free amino acids and glycine betaine in leaf osmoregulation of spinach responding to increasing salt stress


  • Catello Di Martino,

    1. Dipartimento di Scienze Animali, Vegetali e dell’Ambiente, Università del Molise, Via De Sanctis, 86100 Campobasso, Italy;
    Search for more papers by this author
  • Sebastiano Delfine,

    1. Dipartimento di Scienze Animali, Vegetali e dell’Ambiente, Università del Molise, Via De Sanctis, 86100 Campobasso, Italy;
    Search for more papers by this author
  • Roberto Pizzuto,

    1. Dipartimento di Scienze Animali, Vegetali e dell’Ambiente, Università del Molise, Via De Sanctis, 86100 Campobasso, Italy;
    Search for more papers by this author
  • Francesco Loreto,

    1. Consiglio Nazionale delle Ricerche, Istituto di Biochimica ed Ecofisiologia Vegetali, via Salaria Km 29 300 00016 Monterotondo Scalo, Roma, Italy;
    Search for more papers by this author
  • Amodio Fuggi

    Corresponding author
    1. Dipartimento di Scienze della Vita, Seconda Università di Napoli, Via Vivaldi 43, 81100 Caserta, Italy
      Author for correspondence: Amodio Fuggi Tel: +39 0823 274403 Fax: +39 0823 274571 Email:
    Search for more papers by this author

Author for correspondence: Amodio Fuggi Tel: +39 0823 274403 Fax: +39 0823 274571 Email:


  • • The aim of the paper was to determine nitrogen compounds contributing to leaf cell osmoregulation of spinach (Spinacia oleracea) submitted to increasing salt stress.
  • • Sodium, free amino acids and glycine betaine contents were determined in the last fully expanded leaf of plants stressed by daily irrigation with saline water (0.17 M NaCl).
  • • After 20 d of treatment, when Na+ content was c. 55 umol g−1 f. wt above the control, and the reduction of stomatal conductance lowered photosynthesis to c. 60% of the control, the free amino acids of the leaves, especially glycine and serine, strongly increased. Proline and glycine betaine also increased significantly. After 27 d of treatment, when the Na+ content was c. 100 umol g−1 f. wt above the control and photosynthesis was 33% of the control, the free amino acid content, especially glycine and serine, declined. Gycine betaine, but not proline, increased further.
  • • Glycine betaine comprised c. 15% of the overall nitrogen osmolytes at mild salt-stress, but represented 55% of the total, when the stress became more severe. The increase of glycine betaine balanced the decline in free amino acids, mainly replacing glycine and serine (the precursors of glycine betaine) in the osmotic adjustment of the cells. Photorespiration, which increased during salt stress, was also suggested to have a role in supplying metabolites to produce compatible osmolytes.

days of treatment


overall free amino acids


glycine betaine


Salinity and drought are among the major constraints restricting growth and development of plants. In particular, salinity can severely limit crop yield, especially in the most productive areas of the world (Boyer, 1982).

Salt stress involves both osmotic stress, by limiting absorption of water from soil, and ionic stress, resulting from high concentrations of potentially toxic salt ions within plant cells. A variety of protective mechanisms have evolved in plants to allow them to acclimatize to these unfavourable environmental conditions for survival and growth. Although the effect of salinity has been studied in a variety of glycophytic and halophytic plants, the mechanisms of salt tolerance are not well understood (Hasegawa et al., 2000). The synthesis and accumulation of low molecular weight metabolites, known as compatible solutes, is an ubiquitous mechanism for osmotic adjustment in plants. Their main role is to increase the ability of cells to retain water without affecting the normal metabolism (Hamilton & Heckathorn, 2001). Amino acids, sugars, polyols and various quaternary ammonium and tertiary sulphonium compounds may accumulate, as compatible solutes, in many plant species. Betaines, quaternary ammonium compounds occurring naturally in a variety of plants, animals and microorganisms (Rhodes & Hanson, 1993) and proline (Serrano, 1995) are among the most common nitrogen containing compatible compounds. Betaines have been reported to stabilize both the quaternary structure of proteins and membranes against the adverse effect of drought, high salinity, and extreme temperatures (Sakamoto & Murata, 2000). Similar evidence has also been produced for proline (Serrano, 1995). Plants engineered to synthesise such compounds showed an increased protection from photoinhibition or from reactive oxygen species (Hasegawa et al., 2000). In addition, these compatible compounds could serve to store nitrogen that can be used by the plant when the stress is relieved (Rhodes & Hanson, 1993).

In glycophytes, high salinity reduces growth by inhibiting important physiological processes, such as photosynthesis and nutrient supply to the growing tissues, that are directly affected by the reduction of leaf water potential (Munns, 2002).

Spinach is known to be tolerant to mild salinity (Long & Baker, 1986) and to accumulate salt in the leaf cells, but not in the apoplast (Speer & Kaiser, 1991). Like other Chenopodiaceae, spinach accumulates glycine betaine under salt stress (Stewart & Larher, 1980,Weigel et al., 1986).

A previous study on spinach plants irrigated with salt water has shown that a mild salt stress inhibited photosynthesis by restricting the CO2 diffusion because of reduction of stomatal and mesophyll conductance, but did not reduce biochemical and photochemical capacity of the leaves (Delfine et al., 1998). Such leaves were still able to recover their photosynthetic activity when the plants were irrigated with salt-free water (Delfine et al., 1999). The photosynthetic electron transport chain and carbon metabolism, however, were also impaired when the salt stress became severe (Delfine et al., 1998).

The aim of this work was to determine which osmolytes contribute to the osmotic potential in leaves of spinach plants irrigated with salt water. The pattern of free amino acids and glycine betaine in response to the salt accumulation is reported. The rise of the osmotic adjustment and the contribution of nitrogen compounds (amino acids and glycine betaine) under increasing salt stress, are discussed.

Materials and Methods

Plant material and experimental conditions

Two groups of 50 spinach plants (Spinacia oleracea L. cv. Matador) were grown in pots (2 dm3) containing a mixture of soil, peat, and sand (1 : 1 : 1) with a field water capacity of 35%. When five to six leaves were fully expanded, the first group of plants (control) was grown under optimal water conditions by restoring evapotranspiration loss with water. The second group of plants (salt stressed) was irrigated with water containing 0.17 m NaCl to balance the water lost by evapotranspiration. Evapotranspiration was estimated by the pot weight loss before each irrigation. The average daily evapotranspiration from the pots was approx. 15% of field water capacity. All plants were grown in a growth chamber under the same conditions of temperature, light and humidity (day/night temperatures: 27/20°C; light period: 16 h; day/night rh: 50–70%). Measurements of gas exchange were simultaneously taken on ontogenetically similar last well-expanded leaves from both groups after 11, 20, 27, and 43 d of exposure to the treatment (DOT). In conjunction with gas-exchange measurements, leaf discs (3.8 cm2) were cut near the midrib of ontogenetically similar leaves of each group of replicates, frozen in liquid nitrogen, and used to determine the content of salt, amino acids, proline, glycine betaine, chlorophyll a and b and the chlorophyll a : b ratio, as well as Rubisco content and activity. The leaf samples were collected approx. 3 h after the beginning of the light period.

Sodium and potassium determination

Five leaf discs per treatment were dried at 70°C for 24 h and weighed. Sodium and potassium were extracted from each disc in a 10 cm3 mixture of HNO3, HClO4, and distilled water (1 : 5 : 2.5). The solution was kept for 12 h at 100°C, diluted to 25 cm3 with 100 mm HCl, and analyzed by atomic emission spectrometry (I.C.P. Plasma 40, Perkin Elmer, Wellesley, MA, USA).

Amino acid and glycine betaine analyses

Leaf discs (approx. 170 mg f. wt) were ground in liquid N2 to a fine powder that was mixed with 2 ml of ethanol and water (80 : 20 v/v), left for 10 min, collected and centrifuged. The extraction procedure was repeated on the pellet. The supernatants were pooled and used for the analyses. The extraction recovery was always over 95%. The primary amino acids were determined by autosampler-assisted precolumn derivatization by o-phthaldialdehyde (OPA), separation by a high-performance liquid chromatography and detection by fluorescence spectrophotometry with excitation at 340 nm and emission at 450 nm HPLC MT2 system Kontron Instruments: 422 intelligent pump; 482 column thermostat; intelligent autosampler 465; SFM 25 fluorescence spectrophotometer (Kontron, Watford, UK).

The amino acid-OPA derivatives were separated on reverse phase 5 µm C18 ultrasphere column Beckman (250 × 4.6 mm I.D.) kept at 30°C. Solvent A was a mixture of 50 mm sodium acetate adjusted to pH 7 with acetic acid plus 1% (v/v) tetrahydrofuran. Solvent B was pure methanol. The OPA reagent mixture was prepared by mixing 200 µl of K-borate buffer pH 8.5 and 40 µl of OPA dissolved in methanol (5 mg ml−1). Through the autosampler, the extract (100 µl) was derivatized with OPA reagent (50 µl) for 1 min and 20 µl of the mixture were injected and eluted at a flow rate of 1 ml min−1 at 30°C in gradient condition (Di Martino Rigano et al., 1989).

Proline was determined by HPLC as fluorescent 9-fluorenylmethoxycarbonyl derivative (P-FMOC-carbamate). The sample extract was previously derivatised by OPA reagent to remove the primary amino acids. An aliquot (100 µl) of leaf extract was added to 240 µl of OPA reagent and mixed for 1 min. After this, 20 µl FMOC reagent (3 mm in acetone) were added and mixed for 3 min. The excess FMOC reagent was removed by adding 20 µl of heptylamine. The proline was eluted on a reverse phase 5 µm C8 Beckman column (150 × 4.6 mm) in isocratic condition at a flow rate of 0.5 ml min−1. The eluent phase had the following composition: acetate buffer 20 mm pH 7.2 (60%), acetonitrile (20%), methanol (20%). The eluted proline-FMOC-carbamate was detected by fluorescence spectrophotometry set at 254 nm (excitation wavelength) and at 315 nm (emission wavelength).

Glycine betaine was extracted by grinding leaf discs to a fine powder in liquid N2 (about 50 mg f. wt) with 2 ml of distilled water. After incubation for 24 h the homogenate was centrifuged. The extraction procedure was repeated on the pellet. The supernatants were pooled and used for the analyses. The extraction recovery was over 95%. Glycine betaine was determined by HPLC, using a cationic exchange column (Dionex Hypersil SCX, 5 µm, 250 × 4.6 mm) at 30°C in isocratic conditions at a flow rate of 1 ml min−1. The eluent phase was: sodium phosphate buffer 0.05 m, pH 3.7 (95%), methanol (5%) (Kikuchi et al., 1993). Aliquots of the water extract (40 µl) were injected. The eluted glycine betaine was detected by a diode array spectrophotometer set at 195 nm and was characterised and quantified by eluting a standard solution of the pure compound in the same conditions.

Chlorophyll content

Each leaf disc was ground in liquid N2 to a fine powder that was mixed with 2 ml of 80% acetone (v/v). The homogenate, was left in the freezer (−20°C) for 24 h, centrifuged at 10 000 × g at 4°C for 10 min, and the supernatant used to determine chlorophyll content. Chlorophyll a and b contents were determined spectrophotometrically by reading the absorbance at 663.6 and 646.6 nm and using the extinction coefficients and the equations given by Porra et al. (1989).

Measurements of photosynthesis and Rubisco

The gas-exchange system described by Delfine et al. (1998) was used to determine leaf photosynthesis, respiration in the dark, and stomatal conductance. Gas-exchange measurements were taken on five different leaves for each treatment, at a leaf temperature of 25°C. A light intensity of 1200 µE was used when required.

Rubisco measurements were made on leaf discs ground in a chilled mortar with 30 mg of polyvinylpolypyrrolidone, quartz sand, and 2 cm3 of extraction buffer (100 mm Bicine, pH 8.0, 10 m MgCl2, 5 mm DTT, 1 mm EDTA, and 0.02%[w/v] BSA). The solution was centrifuged at 10 000 × g for 10 s.

A fraction of the supernatant was used to radiometrically determine the total carboxylase activity of Rubisco (Di Marco & Tricoli, 1983). Briefly, the assay was conducted at 25°C in vials containing 0.5 cm3 of CO2-free extraction buffer, 20 mm NaHCO3, and 10 mm3 of leaf extract. After 9 min of incubation the radioactive substrate (8.3 kBq with 0.2 µmol NaH14CO3 in 10 mm3 of extraction buffer) was added. One minute later, the reaction was started by adding ribulose-1,5-bisphosphate (1 mm in the reaction mixture) and was stopped after 1 additional min by adding 100 µl of 1 m HCl. The acidified mixture was evaporated to dryness and radioactivity in the residues was measured in a scintillation counter (Packard Instrument Company, Downers Grove, IL, USA).

Another fraction of the supernatant was denatured at 95°C for 5 min in 20% (w/v) SDS, 20% (w/v) mercaptoethanol, and 200 mm Tris-HCl, pH 6.8. Rubisco content was determined in the above solution by SDS-PAGE using a 14% acrylamide gel. Gels were stained with Coomassie Brilliant Blue R-250, destained, and scanned at 550 nm using a Dual-Wavelength Flying Spot Scanner in the transmission mode (model CS-9000, Shimadzu, Tokyo, Japan).

Statistical analyses

The results were consistent in three repeated experiments. In each experiment the measures and analyses were replicated on three different leaf samples derived from different plants for each treatment. The results are mean ± SD of the replicates.

Amino acids, glycine betaine, Na+ and K+ content, were expressed on a f. wt basis for direct comparison.


Sodium accumulation in spinach leaves

A basal concentration of Na+ (around 20 µmol g−1 f. wt) was found in the last fully expanded spinach leaf of control and salt-treated spinach plants in our growth conditions during the first days of culture (Fig. 1). The Na+ accumulation was seen after 10 d of treatment with saline water (10 DOT) and increased linearly until the end of the experiment (Fig. 1). By comparison, the Na+ concentration in the controls did not change significantly during the experiments. The concentration of K+, instead, did not significantly decrease in leaves of salt treated plants compared with controls or change during the experiment (Fig. 1).

Figure 1.

Na+ (closed squares) and K+ (closed circles) of leaves of spinach (Spinacia oleracea) plants irrigated daily with salt water (0.17 M NaCl) during the treatment. Na+ (open squares) and K+ (open circles) contents of plants irrigated with salt free water. data are means ± SD, n= 3.

The leaf d. wt, on a f. wt basis, was similar for both salt-treated and control plants during the experiments, being 10.5 ± 0.7% f. wt. Accordingly, the leaf water content was 89.5 ± 0.7% f. wt.

The decrease of leaf thickness, that did not affect the f. wt, has been reported in spinach under mild salt stress compared with the control (Delfine et al., 1998). It was a result of the reduction of the leaf intercellular air spaces.

Physiological changes in spinach leaves during salt treatment

No difference of photosynthesis occurred in the last fully expanded spinach leaves of treated plants compared with the controls during the first 10 DOT (Table 1). Growth and leaf expansion were also not inhibited (not shown). Photosynthesis decreased when the salt began to accumulate in the leaves (Table 1). At 20 DOT, when Na+ accumulated in the leaves was 76.8 µmol g−1 f. wt, photosynthesis as well as stomata conductance was approx. 60% of the control. At the same time, the activity and protein content of Rubisco, as well as chlorophyll content and chlorophyll a and b ratio, were unchanged (Table 1).

Table 1.  Na+ concentration, photosynthesis, stomatal conductance, Rubisco activity and content, chlorophyll (a + b) concentration in spinach (Spinacia oleracea) leaves of plants irrigated with water (C) and salt water (0.17 m NaCl) (T)
Days of treatment (d) Sample10202743
  1. SD in parenthesis.

(µmol g−1 f. wt)
(µmol m−2 s−1)
(mol m−2 s−1)
Rubisco acticity
(µmol m−2 s−1)
Rubisco content
(µmol m−2)
(a + b)
(µmol m−2)

The decline of photosynthesis in leaves of mild stressed spinach plants was mainly a result of reduced stomata and mesophyll conductances (Delfine et al., 1998).

Subsequently, negative effects of salt accumulation on the photosynthetic apparatus were observed. At 27 DOT, when the Na+ content in salt stressed leaves was 117 µmol g−1 f. wt, photosynthesis and Rubisco activity and content were reduced to 33%, 42% and 54% of the control values, respectively. Stomatal conductance was approx. 40% of the control and did not change significantly (Table 1). In addition, after 20 DOT, apical growth and leaf expansion were strongly reduced (not shown) and, so, the last fully expanded leaves were the same as at 20 DOT. At 43 DOT, when the Na+ content of salt stressed leaves was approx. 210 µmol g−1 f. wt, photosynthesis and Rubisco activity and content were approx. 20% and 30% of control values, respectively (Table 1).

Chlorophyll content, as Rubisco, was not affected by salt until 20 DOT. It, then, decreased significantly in salt stressed leaves. It was approx. 50% and 30% of the control values at 27 and at 43 DOT, respectively (Table 1).

Free amino acids and glycine betaine during salt treatment

The leaf pool of free amino acids did not differ significantly between controls and salt stressed plants in the early DOT (not shown). At 10 DOT the amino acid distribution and content was still similar in the last fully expanded leaves of salt stressed and control plants, as for the leaf content of Na+ (Table 2). The most abundant free amino acids were glutamate, glutamine, serine, glycine, alanine and aspartate. The pool of other amino acids (valine, leucine, isoleucine, lysine, tryptophan and proline) was approx. 1% of the overall amount. Only traces of glycine betaine were found.

Table 2.  Concentration (µmol g−1 f. wt) of free amino acids, glycine betaine and Na+ in leaves of spinach (Spinacia oleracea) plants irrigated with water (C) and salt water (0.17 M NaCl) (T)
Days of treatment (d) Samples10202743
  1. SD was always lower than 12% of the average value.

Glycine 0.63 0.88 0.54 8.99 0.61  2.52 0.50  0.79
Serine 0.78 1.02 0.68 3.40 0.61  1.28 0.38  0.66
Proline 0.11 0.14 0.08 2.16 0.07  2.51 0.05  0.78
Aspartate 0.56 0.57 0.38 0.24 0.32  0.09 0.14  0.12
Glutamate 2.42 3.33 1.67 0.37 1.48  0.24 0.95  0.25
Asparagine 0.45 0.35 0.18 0.13 0.15  0.03 0.07  0.06
Arginine 0.84 0.68 0.52 0.23 0.51  0.45 0.34  0.32
Glutamine 1.17 1.17 0.81 1.00 0.74  0.47 0.77  0.45
Alanine 0.55 0.60 0.38 0.34 0.43  0.36 0.42  0.17
Tryptophan 0.05 0.07 0.03 0.02 0.02  0.04 0.06  0.23
Valine 0.04 0.04 0.03 0.02 0.02  0.03 0.06  0.26
Isoleucine 0.04 0.05 0.02 0.03 0.03  0.04 0.04  0.20
Leucine 0.05 0.04 0.03 0.03 0.03  0.08 0.05  0.24
Lysine 0.06 0.07 0.06 0.05 0.05  0.06 0.07  0.33
Σ AA 7.75 9.04 5.6517.04 5.39  9.07 4.26  8.79
Glycine betaine 0.02 0.02 0.03 2.25 0.02 11.00 0.25  3.25

The pattern of amino acids and glycine betaine changed significantly in leaves of salt-stressed plants at 20 DOT (Table 2). Glycine and serine content strongly increased compared to controls (17- and fivefold, respectively). These two amino acids accounted for 68% of the overall amino acid content of salt stressed leaves, whilst they accounted for < 20% in the control. Among the other amino acids, glutamine and alanine slightly increased with respect to the controls. Proline also increased (30-fold) compared with the controls and accounted for > 17% of the overall free amino acid content. Glutamate, aspartate, asparagine decreased. The remaining amino acids showed no significant changes. The overall free amino acid content (Σ AA) of salt stressed leaves was 300% higher than in the controls, mainly because of the strong increase in glycine, serine and proline, that contributed 85% of the total (Table 2). Glycine betaine also increased (approx. 20-fold over the control).

At 27 DOT, the leaves of salt-stressed plants showed a further change in amino acid pattern. Their overall amino acid content decreased at 180% of that found in the controls. Glycine and serine decreased strongly, but their content was still significantly higher than in controls and accounted for 41% of the overall free amino acid content (Table 2). Proline, on the other hand, increased further (10% more than at 20 DOT) in those leaves, accounting for 39% of the overall amino acid content.

At 27 d, glutamine, glutamate, and aspartate still decreased compared with 20 DOT in stressed leaves. Among the most abundant amino acids, alanine did not change. Leucine, isoleucine, valine and tryptophan also did not change significantly. Their concentrations were still similar to the control leaves and contributed to the overall amino acid content for < 3%. Glycine betaine, on the other hand, still increased strongly reaching approx. 500% of the amount found in 20 DOT stressed leaves.

At 43 DOT, the overall free amino acid content further decreased to 200% of the controls in salt stressed leaves. Nevertheless, the amino acid distribution was very different in the two treatments (Table 2). Glycine and serine concentrations decreased to a value similar to the controls. Proline decreased to about one third of the 20 DOT stressed leaves, but its content was still 10-fold higher than the controls. Glutamine, glutamate, aspartate and asparagine contents were lower than in the controls. On the other hand, leucine, isoleucine valine and tryptophan increased on average fivefold with respect to the 27 DOT stressed leaves, suggesting that a large protein degradation had taken place, in agreement with the large loss of Rubisco (Table 1). The salt stressed leaves turned yellow, appeared clearly aged and their physiological performance was much lower than the controls (Table 1). Glycine betaine also decreased to about one third of the content of 27 DOT stressed leaves.

The amino acid content in the last fully expanded leaves of control plants slowly declined during the experiment (Table 2). However, a similar pattern of amino acids and glycine betaine was observed at 20 and at 27 DOT. At 43 DOT aspartate, glutamate, serine, glycine, asparagine, glutamine were among the amino acids whose content decreased. Alanine and the other amino acids (leucine, isoleucine, valine, tryptophan, lysine and proline), as glycine betaine, did not change (Table 2).

Osmolyte distribution in spinach leaf cells

To balance the osmolarity (osmolality) increase from salt accumulation in the vacuoles, the cell could confine the organic osmolytes to the cytoplasm and the other hyaloplasm organelles (Robinson et al., 1983; Schroeppel-Meyer & Kaiser, 1988).

Analysing the volume distribution in spinach leaves by the stereological method, Winter et al. (1994) reported that the overall aqueous (apoplast + hyaloplasm) and hyaloplasm volumes in the leaf tissues accounted for 89 ± 1% and 79 ± 3% of the leaf f. wt, respectively. In addition, at the subcellular level, the vacuolar volume was 88 ± 4% of hyaloplasm volume, while the remaining 12 ± 4% accounted for the (chloroplast, cytoplasm, mitochondria and nucleus) volume.

These volume data were used to estimate the osmolarity increase of vacuole and the other hyaloplasmic organelles of leaf cell of salt stressed plants compared with the controls, at different DOT.

The osmotic potential increase of leaves of spinach plants submitted to salt stress was largely a result of accumulation of Na+ and Cl within the leaf cells (Robinson et al., 1983). A decrease of K+ content was observed in such experiments; however, the overall increase of (K+ + Na+) balanced the increase of Cl.

Such ions largely accumulated in the vacuole. They accounted for over 94% of the leaf contents on a f. wt basis (Speer & Kaiser, 1991).

In our experiments the salt treatment did not significantly decrease the leaf K+ (Fig. 1), suggesting that the content of Cl and Na+ were equal.

Therefore, the overall osmolarity increase of vacuole compared with the control was estimated, by doubling the Na+ concentration, calculated considering that the 94% of the leaf sodium content on a f. wt basis was in the vacuole.

The overall osmolarity increase of cytoplasm and organelles, as a result of nitrogen compounds, was calculated by summing those of the overall amino acids and Glycine betaine (Σ AA + GB; see Table 2) and subtracting those of the control, considering that glycine, serine, proline, and glycine betaine, the main contributors to the osmolarity increase, have no net charge at the cytoplasm and organelle pHs.

In this analysis no assumption was necessary for the distribution among the organelles of the various contributing compounds.

Figure 2 shows that at 20 and 27 DOT the NaCl-dependent osmolarity increase in the vacuole of salt-stressed leaf cell was nearly equal to that provided by the overall increase of amino acids (Σ AA) and Glycine betaine (GB) in the cytoplam and organelles. However, these nitrogenous osmolytes changed in this period. The main contribution at 20 DOT was given by the amino acids glycine, serine and proline, while at 27 DOT it was by glycine betaine.

Figure 2.

Estimated osmolarity of salt (NaCl (open columns)) in the vacuole and of free amino acids (Σ AA (grey columns)) and glycine betaine (GB (black columns)) in the other hyaloplasm compartments of leaf cells of spinach (Spinacia oleracea) plants irrigated with salt water (0.17 M NaCl) with respect to controls irrigated with water.

At 43 DOT, these compounds contribute no more than 10% to the osmotic adjustment, suggesting that other osmolytes were involved or that a large loss of living cells was taking place.


Mild salt stress

After 10 DOT, sodium enhancement could be detected in the last fully expanded leaves of salt-stressed spinach plants. The strong increase of glycine and serine (Table 2), at 20 DOT, when the Na+ level in the leaf was 75 µmol g−1 f. wt, suggested that the early effect of salt was an increase in the photorespiration rate (Di Martino et al., 1998). Both amino acids are strongly involved in the photorespiratory cycle (Lea et al., 1990). Glycine is produced by the transaminase between glyoxylate and glutamate, and is the substrate for the glycine decarboxylase: serine hydroxymethyltransferase complex, that, in turn, produces serine, ammonia, and carbon dioxide.

On the other hand, photosynthesis and stomatal conductance were reduced, even if no significant change occurred in Rubisco activity and content as well as in chlorophyll content (Table 1), supporting the idea that photorespiration increased under salt stress.

In agreement with this suggestion, gas exchange and fluorescence studies in spinach exposed to mild salt stress showed that the estimated CO2 concentration within the chloroplast was lower than that of the control plants, supporting the view that photosynthesis was restricted by the decrease of mesophyll and stomata conductance (Delfine et al., 1998). The electron transport rate and rubisco activity, in fact, did not suffer significant changes.

Leaves suffering from a mild salt stress were still able to recover their photosynthetic activity when the plants were irrigated with salt-free water, confirming that the photosynthetic structural components were not irreversibly damaged (Delfine et al., 1999).

Among the other amino acids, the reduction of glutamate can be explained by its utilization as substrate for glutamate: glyoxylate aminotransferase and glutamine synthetase. These enzymes are involved in the synthesis of glycine and in the recycling of ammonia produced by glycine decarboxylase, respectively. In such a view glutamine should increase, and aspartate and asparagine should decrease, being drained through glutamate towards glycine (Lea et al., 1990). Glutamate was also involved in the synthesis of proline.

The increase of alanine in stressed leaves suggests that glycolysis and thus, respiration were increased to sustain the higher energy demand to confine salt to the vacuoles and to furnish carbon skeletons for the photorespiratory cycle (Lea et al., 1990).

Leucine, isoleucine, valine, tryptophan and lysine contents remained similar in salt stressed and control leaves, supporting the view that no change occurs in protein synthesis or degradation when salt accumulation is mild (Robinson et al., 1992).

Proline synthesis was also induced in spinach plants submitted to mild salt stress (Di Martino & Fuggi, 2001a), as already reported in many plant species (Stewart & Larher, 1980). Its concentration in the stressed leaves was 30-fold higher than in the controls, accounting for 18% of the overall free amino acid content at 20 DOT. Similar results were also found in rosette leaves and in roots of Arabidopsis subjected to water and salt stress, respectively (Chiang & Dandekar, 1995). Proline reached concentrations eight- to 10-fold over the control for plants kept in hydroponics with NaCl, and a higher increase was observed when plants were kept in a nutrient solution supplemented with polyethylene glycol.

Spinach is known to be an accumulator of glycine betaine under stress. However, in the early onset of stress, at 20 DOT, glycine, serine and proline contributed to 85% of the overall amount of the nitrogenous osmolytes, glycine betaine having only a minor role.

Since the equivalent osmolarity of salt in the vacuole was similar to that of the overall amino acids and glycine betaine in the other organelles (Fig. 2), other compatible solutes, such as sugars, seemed to play only a minor role.

This is supported by the view that the overall increase of (Na+ + K+) and Cl concentrations on a leaf water basis accounted for approx. 83% of the overall leaf osmotic potential increase in spinach plant cultured under salt stress conditions (Robinson et al., 1983). However, considering that the vacuole water volume is approx. 80% of the overall leaf water volume (apoplast + hyaloplasm), and the overall increase of (K+ + Na+) and Cl content is largely confined to the vacuole (Speer & Kaiser, 1991), their concentration increase could attain an osmotic potential increase equal to the value measured for the whole leaf.

In this view, the increase of osmolarity of the cytoplasm and the other hyaloplasm organelles, supported by the nitrogen compounds (Fig. 2), could be sufficient to adjust the cytoplasm and the other intracellular organelles to the same osmotic potential of the vacuole.

Severe salt stress

At 27 DOT, sodium concentration in the salt stressed leaves ([Na+] = 150 µmol g−1 f. wt) was sixfold higher than in controls and severely inhibited cell metabolism. The further decrease of photosynthesis (the residual activity was about 30% of the control) was associated with inhibited biochemical and photochemical capacities of the leaf: rubisco activity and content and chlorophyll content were approx. 50% of the control (Table 1; Delfine et al., 1998).

Glycine and serine, as well as the overall free amino acid pool decreased to 50% of the concentration found at 20 DOT. Glycine and serine were still the most abundant free amino acids (glycine and serine accounted for 45% and 25% of the overall amino acid amount, respectively). Proline concentration did not change significantly.

However, at this stage glycine betaine strongly increased reaching a concentration exceeding the overall amino acid content in the salt stressed leaves, suggesting that, when the stress was severe, glycine betaine substituted glycine and serine as an osmolyte.

Glycine betaine in spinach plants, has been reported to derive from the oxidation of choline, that in turn, derives from serine through ethanolamine (Summers & Weretilnyk, 1993). Its synthesis can derive from serine and ultimately, from glycine, intermediate compounds of the photorespiratory cycle. Therefore, it is possible that, at least in salt stressed spinach leaves, photorespiration not only provides a sink for photosynthetic electron transport, dissipating the excess of photochemical energy when the dark reaction activity is reduced, but also mobilizes carbon reserves through the glycolate oxidation pathway, contributing to the synthesis of compatible nitrogen solutes for plant acclimation (Di Martino & Fuggi, 2001b).

Proline did not increase at 27 DOT, still giving only a minor contribution as osmolyte. However, it has been reported that glycine betaine reduced proline accumulation in spinach (Sulpice et al., 1998).

At 43 DOT, when the sodium concentration in the salt stressed leaves was 224 µmol g−1 f. wt (Table 2), all the amino acids involved in the osmotic adjustment decreased further, as did glutamate, aspartate, asparagine, glutamine and alanine. Glycine betaine concentrations also decreased. The significant increase in the amount of leucine, isoleucine, valine, lysine and tryptophan suggests that protein degradation occurs, as in aged leaves before abscission or in plant cell resting cultures (Stewart & Larher, 1980; Robinson et al., 1992). Their increase therefore may indicate that severely salt-stressed spinach leaves were suffering from premature ageing.

Osmotic adjustment in spinach leaf cell

Plant cells, kept in a low water potential environment, increase the internal concentration of osmolytes. To avoid shrinkage and/or swelling of intracellular organelles, equivalent concentrations of such compounds must be kept in the intracellular compartments, irrespective of the osmolytes found in each organelle.

Spinach cells store salt ions in the vacuoles (Speer & Kaiser, 1991) and other osmolytes in the cytoplasm and in the other hyaloplasm organelles to osmotically balance them with respect to vacuoles. The similar increase of osmolarity as a result of salt stored in the vacuole and organic nitrogen compounds (amino acids and glycine betaine) in the cytoplasm and other hyaloplasm organelles at 20 and at 27 DOT, suggests that such a distribution can achieve this balance (Fig. 2).

Amino acids and glycine betaine were in low concentrations within the spinach cell vacuoles under salt stress (Schroeppel-Meyer & Kaiser, 1988; Winter et al., 1994). Moreover, differential distribution of the organic osmolytes in the salt stress and water stress are suggested by data from rape, a species accumulating proline (Larher et al., 1996): rape leaf discs synthesized proline at amounts 10-fold lower when they were exposed to a NaCl solution than when exposed to polyethylene glycol 6000 of the same water potential). Such data fit in with the hypothesis that in polyethylene glycol treated leaves, proline accumulated in all cell compartments, while in NaCl treated ones the salt was confined to the vacuole, and proline to the cytoplasm and the other hyaloplasm organelles. Osmotic equilibrium within the cell was attained if the vacuole, on the one hand, and cytoplasm and organelles, on the other hand, accounted for 90% and for 10% of the overall cell volume, respectively.

A slight increase of amino acid content and, in particular proline, over the controls has also been reported in spinach leaves exposed to salt stress compared with those exposed to water stress (Sulpice et al., 1998).

Accumulation of the salt ions in the vacuole to attain the cell osmotic balance under salt stress is a successful mechanism. Cells, in fact, strongly reduce their need to invest valuable metabolites in the synthesis of organic osmolytes.

In conclusion, a mild salt stress modifies the pattern of free amino acids, strongly increasing the level of glycine and serine. This suggests that the flux of photorespiratory carbon had increased. Photorespiration, that provides a sink for photosynthetic electron transport when photosynthesis is reduced, may also be involved in the production of compatible solutes for the control of salt stress.

The osmolytes that support the increase of the cell osmotic potential under salt stress are mainly nitrogenous compounds (glycine, proline, and glycine betaine), but their synthesis is not synchronous: glycine, serine and proline contributed, early, at the onset of the stress, while glycine betaine contributed mainly when the stress became severe. However, when salt accumulation in the leaves became very high, such compounds no longer seemed to control the increase of osmotic potential.

The osmolytes synthesised in the salt stressed cells are confined to the cytoplasm and other hyaloplasmic organelles, in contrast with drought stress when they are also released in the vacuole.


The research was supported in part by ‘Ministero dell’Istruzione, dell’Università e della Ricerca’ and by ‘Consiglio Nazionale delle Ricerche’ of Italy. Thanks are especially due to Giuseppe Palumbo for his assistance in determining cations in the plant material.