Conventional crop and pasture breeding for salt tolerance

As described in the companion article (Salinity Stress and its Impact), species vary in their capacity to tolerate salinity. Amongst the major crop species, barley, cotton, sugar beet and canola are the most tolerant; wheat and lucerne (alfalfa) are moderately tolerant, while rice and most legume species are sensitive. For a more complete list of crops see Richards (1969).

There is also variation in salinity tolerance within species, especially in outcrossing species like lucerne. The classical breeding approach is to (1) screen collected germplasm for salinity tolerance, (2) cross the identified tolerant types with the desired cultivars and (3) selecting the desired plant types having salinity tolerance as well as other desirable traits from the advancing and segregating generations. By exploiting the naturally occurring genetic variability that exists within a species, some relatively tolerant cultivars have been developed for crops including rice, wheat, lucerne, white clover and citrus.

There is probably a great diversity in salinity tolerance within species that has not been fully explored and exploited. One reason for this is the difficulty of screening large numbers of individuals for small, repeatable and quantifiable differences in biomass production, let alone yield. Obtaining a wide range of germplasm with potential genetic differences in salinity tolerance is not difficult, because international collections are usually available to any scientist. However, the difficulty lies in how to measure salinity tolerance.

How to measure salinity tolerance:

Salinity tolerance is difficult to measure because of its complexity. Not only are there a number of genes controlling salinity tolerance whose effect interacts strongly with environmental conditions, but there are two major and distinct components of salinity tolerance which can often be difficult to distinguish (Munns 2002).

Salinity imposes two major stresses on the plant: one is a high osmotic pressure in the soil solution making it harder for the plant to extract water, the second is a high NaCl concentration in the soil solution that makes it difficult for the plant to exclude the NaCl while taking up other cations and anions. The effect of these two stresses are seen in sequence. Salinity lowers the water potential of the roots, and this quickly causes reductions in growth rate, along with a suite of metabolic changes identical to those caused by water stress. Later, there may be salt-specific effects that impact on growth or senescence. This two-phase model is summarized in Fig. 1, and described in more detail in the companion article ‘Salinity Stress and its Impact’.

How to screen for differences in salinity tolerance within species based on growth:

Screening based on growth needs to allow for the two distinct mechanisms for salinity tolerance: tolerance to the osmotic effect of the saline soil solution, and tolerance to the salt-specific nature of the saline solution.

The osmotic effect alters the water relations of the plant, and reduces the rate of cell expansion. This leads to a reduction in the rate of development of new roots, leaves and lateral shoots. The osmotic effect also reduces stomatal conductance, which leads to reduced photosynthesis. It also causes accelerated senescence of older leaves. Thus there are three somewhat independent processes being affected (i.e. new leaf formation, old leaf death and photosynthetic activity) that all contribute to a reduction in the net assimilation rate of the plant. These effects are identical to drought.

The osmotic effect occurs instantly the soil water potential decreases, and recovers instantly it increases (Fig. 2). If the period of stress is short (hours) the recovery is complete. If the period of stress is long, the recovery is more limited, as stress may have already reduced the number of lateral shoots, and the number of cells in the dividing zones of growing roots and leaves, so that there are a reduced number of cells to respond.

Responses of leaf elongation are greater than of root elongation (Munns 2002). The reason for this is not known, but it means that leaf elongation or leaf expansion rates are a more sensitive indicator of osmotic stress than root elongation rates.

The genetic variation in the growth response to the osmotic effect of salinity is likely to be small, both within a species, and across similar species. For example, we found little difference between the effect of salinity on leaf elongation in 15 cultivars of bread wheat, durum wheat, barley and triticale, cultivars with established differences in reputation for salinity tolerance. All 15 cultivars had the same decrease when the salt was increased from 0 to 250 mM NaCl over 10 d. Even the most salinity-sensitive genotype, a durum wheat cultivar, had the same percentage reduction as the most salinity-tolerant genotype, a barley cultivar. In fact, there is a remarkable similarly between different species in the osmotic response in saline solution. For instance, 100 mM NaCl causes approximately 50% decrease in leaf elongation rate of the salinity-sensitive species maize and rice, and nearly as much in the salinity-tolerant species bread wheat and barley. Longer term growth responses clearly differ.

The osmotic effect has the same characteristics as drought, so genotypes could be screened for drought tolerance. But this is not easy either (see 'Drought Stress and its Impact).

The salt-specific effect causes increased uptake of Na+ and Cl-, and decreased uptake of essential cations particularly K+ and Ca2+. Uptake of Cl- may alter uptake of anions such as phosphate and nitrate but these effects are complex and vary between species. If the uptake of Na+ or Cl- exceed the plant's ability to partition the ions between different tissues or organs, or to sequester ('compartmentalise') the ions within the cell's vacuole, these ions build up in the cytoplasm to toxic concentrations.

The salt-specific effect shows up in old leaves that have accumulated excessive amounts of Na+ and Cl-. While it may be visible, in that old leaves die earlier, it may not affect the rate of new leaf production for some time. The salt-specific effect on growth is not seen until after the osmotic effect on growth (Fig. 1). The length of time required before the growth differences between genotypes due to the salt-specific effect can be seen depends on the salinity and the degree of tolerance of the species. This represents a 'second phase' of the growth reduction. The second phase will start earlier in plants that are poor excluders of Na+, and will start earlier when root temperatures are higher. For plants such as rice that are grown at high temperatures, two weeks days in salinity is sufficient to generate differences in biomass between genotypes that correlate well with differences in yield. However for temperate crops, as at least a month is needed for genotypic differences in the response of biomass production to take effect.

The labour and space demands of these long experiments makes them impractical for screening large numbers of genotypes, or selecting salinity-tolerant progeny resulting from crosses with cultivars. Not only do plants need to be grown for lengthy periods of time, but controls need to be included. Control plants consumes a large area of glasshouse space if they are to be grown at their optimum rate, with sufficient space so that radiation is not limiting. In the field, the major drawback is the heterogeneous nature of soil salinity and other soil constraints.

Screening for a trait associated with a specific mechanism is preferable to screening for salinity tolerance itself. Screening for specific traits can reduce the time needed to grow plants in salinity, and can eliminate the need to grow plants under control conditions, thus making savings on glasshouse space and labour.

The importance of traits:

Because of the complex nature of salinity tolerance, as well as the difficulties in maintaining long-term growth experiments, we recommend trait-based selection criteria for screening techniques. Specific traits are less subject to environmental influence than growth rates.
Further, this allows for different traits to be pyramided, especially when molecular markers for specific traits have been identified.

The most successful trait in terms of plant breeding relates to rates of Na+ or Cl- accumulation in leaves, measured as the increase in salt in a given leaf over a fixed period of time. This trait has a high heritability and has been used to develop cultivars of rice, white clover, and lucerne with increased tolerance to saline soil. Sometimes K+/Na+ discrimination instead of Na+ exclusion is used for screening, however the uptake of K+ and the resultant K+/Na+ discrimination may be the result of genetic differences in the regulation of Na+ uptake, and not independent of it. If so, there is nothing to be gained by measuring K+ as well as Na+.

A correlation between Na+ or Cl- accumulation and salinity tolerance is found in most species, however not all species contain significant genetic variation in Na+ or Cl- accumulation. Durum wheat is one species in which there is significant genetic variation in Na+ but not Cl- uptake. Genotypes with low Na+ uptake are more salinity tolerant, as indicate by greater biomass production over 1-2 months in saline solution, in comparison to non-saline solution (Fig. 3). In some species, Na+ is retained in roots and stems in exchange for K+, and only Cl- progresses through to the leaves, balanced by K+. In those cases, Cl- exclusion correlates with salinity tolerance.

Proving that Na+ (or Cl- ) exclusion confers salinity tolerance in terms of yield is not so easy. A comparison between durum landraces with very low and very high rates of Na+ accumulation showed that, at moderate salinity, the yield of genotypes with low Na+ was greater than those with high Na+, but at high salinity there was no yield advantage. The osmotic effect of the salinity then dominated (Fig. 4).

Tissue tolerance

Tissue tolerance, i.e. tolerance of high internal Na+ concentrations, cannot be measured directly, and is difficult to quantify. Yet it is clearly important; overexpression of the vacuolar Na+/H+ antiporter that sequesters Na+ in vacuoles (NHX1) improved the salinity tolerance of Arabidopsis, tomato and brassica (see next section).

Tolerance of high internal Na+ levels is evidenced by an absence of leaf injury despite high leaf concentrations of Na+. Concentrations of Na+ above 100 mM will start to inhibit most enzymes, so when tissue concentrations are over 100 mM, which corresponds to about 0.5 mmol g-1 DW (assuming a leaf water content of 5 g H2O g-1 DW), the Na+ must be compartmentalised in vacuoles, and be a higher concentration there than in the cytoplasm.

compartmentation Na+ in vacuoles to some extent, as levels of Na+ up to 1 mmol g-1 DW (200 mM) are quite common in photosynthetically active leaves of many species. In a study in wheat genotypes, Na+ became potentially toxic only when leaf concentrations exceeded 1.25 mmol g-1 DW (250 mM), as judged by the onset of non-stomatal reductions in photosynthesis in durum wheat at this concentration (James et al, 2002).

There may be genetic variation for tolerance of high internal Na+ concentrations in many species, as indicated by 'out-lyers' in correlations between leaf Na+ concentrations and salinity tolerance. This has been found in rice and wheat and probably many other species. However, quantitative assess of the genetic variation is difficult. The trait is characterised by leaf longevity, lack of necrosis, and prolonged growth despite very high accumulation of Na+. A recent paper describes the success or failure of various methods to screen durum wheat for genetic variation in tolerance of high internal Na+ (Munns and James 2003). The degree of leaf death was measured as an indicator of Na+ toxicity. Leaf injury, however could arise from a number of reasons. First there would be the osmotic effects of salt in the soil solutions, causing accelerated senescence due to leaf water deficit or hormonal effects arising from root signals. Second, there could be nutrient imbalances resulting in deficiencies or excesses of other ions. Third, there could be dehydrating effects of salts building up in the cell walls. A phenotype other than leaf injury is needed to distinguish between genotypes with different degrees of tissue tolerance. The onset of non-stomatal reductions in photosynthesis mentioned above is a possible phenotype.

Traits for the osmotic effect:

Measurements of growth, survival, leaf gas exchange and germination will indicate the osmotic effect of salinity. If the experiments are conducted for lengthy periods of time, measurements of growth, survival and gas exchange will also reflect toxic effects of salts in the leaves.

Growth can be measured as leaf elongation, root elongation, leaf area expansion, or shoot biomass. Of these indices of growth, leaf area expansion or shoot biomass are the most sensitive and comprehensive, as this includes production of tillers or lateral shoots. The number of lateral shoots is more sensitive to water stress than the elongation rate of any given leaf.

Gas exchange characters are stomatal conductance (including infra-red temperature), photosynthesis, transpiration efficiency (including carbon isotope discrimination), and chlorophyll fluorescence. For screening larger numbers of genotypes, traits related to stomatal conductance are the most feasible. Chlorophyll fluorescence as an indicator of photosynthetic activity was assessed by James et al (2002). They found that the simplest parameter, Fv/Fm, was no more sensitive than chlorophyll content itself. The quenching parameters, such as NPQ, did indicate a reduction in photosynthetic capacity before chlorophyll itself started to degrade (James et al. 2002), but these parameters require both light and dark-adapted readings and are not feasible for large numbers of genotypes.

Leaf injury can be quantified by a number of methods such as chlorophyll content, or electrolyte leakage of cut discs. Chlorophyll content can be measured with a SPAD meter.

Turgor and osmotic adjustment are also indicators of leaf injury and photosynthetic capacity.

All these methods are used also for drought tolerance.

Whole plant survival at high salinity may reflect the salt-specific effect rather than the osmotic effect, as the plant's ability to control salt uptake may break down at high salinity.

Germination is easy to measure, but little or no relation between salinity tolerance at germination and that of the seedling or adult plant has been found in any species examined.