Correlation between salt tolerance and genetic diversity between Sulla carnosa and Sulla coronaria

Sulla constitutes an important genetic resource contributing to pastoral production, particularly in semi-arid regions. In Tunisia, seedlings of the southern species Sulla carnosa (Desf.) and the northern species Sulla coronaria (L.) were treated with NaCl (0, 100 and 200 mM) for 27 days. Salt treatments decreased leaf dry matter more in S. carnosa than in S. coronaria . S. coronaria accumulated less Na + and greater amounts of K + and showed greater K/Na selectivity, a trait which could be related to the maintenance of higher net K + uptake and transport in the presence of NaCl. Pigments were severely affected by salt stress in leaves of S. carnosa when compared with leaves of S. coronaria . In addition to these physiological characterisations, genetic diversity was measured between the two accessions using inter simple sequence repeat (ISSR) markers. Three ISSR primers generated a total of 63 DNA amplicons for S. carnosa and 64 DNA amplicons for S. coronaria , all of which were polymorphic between the two accessions. Correlations between the molecular and physiological data revealed statistically significant correlations between the salt response of these two Sulla accessions and two molecular markers B340 and B860, in roots and shoots, respectively. S. coronaria showed greater salt tolerance on the basis of growth and K/Na selectivity, making it a good candidate for inclusion in a future breeding programme. fragment (band) was scored as 1 when present and 0 when absent. Bands showing unambiguous polymorphism between salinity treatments and/or S. carnosa and S. coronaria were entered into a binary data matrix. Principal components analysis (PCA) was conducted on the ISSR patterns using XLSAS software (Statistical Analysis System, version 2009.3) (SAS, 2009). A Pearson correlation test and a Mantel test (Mantel, 1967) were used to test the statistical relationship between the presence or absence of specific ISSR amplicon patterns and the elements of the two matrices generated by the genetic, growth and physiological traits.

#These author contributed equally to this to this work.1981).Plant nutrition depends on the activity of membrane transporters that translocate minerals from the soil into the plant and mediate their intra and intercellular distribution (Epstein and Bloom, 2005;Marschner, 1995).High amounts of Na + in the growth medium perturb nutrient uptake in plants (Marschner, 1995;Grattan and Grieve, 1999;Ashraf, 2004).Selective ion uptake and differential ion compartmentalization are main features that explain disparities in tolerance to salinity between glycophytes and halophytes (Greenway and Munns, 1980).Na + compartmentalization is achieved by the action of Na + /H + antiporters.The Arabidopsis AtNHX1 protein was the first Na + /H + exchanger identified in plants (Gaxiola et al., 1999) and it mediates both Na + and K + coupled transport in vacuoles (Venema et al., 2002).Salinity affects nutrient acquisition by interfering with K + uptake by ion carriers and ion channels (Maathuis and Amtmann, 1999).In Arabidopsis thaliana, passive uptake of K + into roots is mediated by K + selective channels located on the plasma membrane of root cells (Hirsch et al., 1998;Kim et al., 1998;Broadley et al., 2001;Gierth et al., 2005).Long distance transport of K + involves AKT2 channels, which secrete K + into the phloem in shoots (Lacombe et al., 2000) and SKOR, a shaker like outward channel which releases K + into the xylem of roots (Gaymard et al., 1998).The reduction in uptake of mineral nutrients (such as K + ) under saline conditions may occur due to competition with Na + .However, maintaining a high cytosolic K + /Na + ratio is a criteria of salt tolerance in species, such as barley and wheat, when they are confronted with a saline medium (Cuin et al., 2003;Dvorak et al., 1994;Gorham et al., 1990Gorham et al., , 1997;;Rubio et al., 1999).In A. thaliana, salt tolerance of ecotype NOK2 also was related to a higher K + /Na + selectivity (Kaddour et al., 2009).
Salinity modifies the level of secondary metabolites such as carotenoids (CAR) and chlorophylls (Agastian et al., 2000;Parida et al., 2002;Mahmoudi et al., 2010).Salt tolerance has been correlated to the maintenance of the level of these metabolites and the transcription of their genes (Mahmoudi et al., 2010).Indeed, steady levels of CAR are strongly correlated to salt tolerance of mungbean plants (Wahid et al., 2004) and sugarcane cultivars.In sugar cane, carotenoids may constitute a strategy for achieving salt tolerance along with limitations to the reduction of chlorophyll (Kanhaiya, 1996).
Sulla is a group of leguminous plants represented by six species of the Sulla and Hedysarum genii, which are spread across the mediterranean basin (Trifi-Farah et al., 2002;Choi and Ohashi, 2003).Accessions of Sulla are exploited for protection and enrichment of soil due to their N 2 fixation properties.Hence, Sulla constitutes an important genetic resource and contributes to pastoral production particularly in semi-arid regions because of its drought tolerance (Trifi-Farah and Marrakchi, 2002).
PCR-based molecular marker approaches are in demand when studying natural biodiversity because of their simplicity and requirement for only small quantities of sample DNA.Among PCR-based marker methods, AFLP (amplified fragment length polymorphism) (Karp et al., 1997), RAPD (random amplified polymorphic DNA) (Williams et al., 1990), SSR (simple sequence repeats or microsatellites) (Karlin and Burge, 1995) and ISSR (inter simple sequence repeats) (Zietkiewicz et al., 1994) are most commonly used.Genetic diversity between a number of Sulla accessions was revealed using AFLP, RFLP and rDNA ITS sequence variation (Trifi-Farah and Marrakchi, 2002;Marghali et al., 2005;Chennaoui et al., 2007).However, ISSRs have never been used to definite genetic variation between Sulla accessions.These PCR-based molecular markers involve amplification of DNA by a single 16 to 18 bp primer composed of a repeated non anchored sequence or a repeated sequence anchored at the 3' or 5' end by 2 to 4 arbitrary nucleotides (Zietkiewicz et al., 1994).ISSR markers are easy to handle, highly informative and repeatable.
Sulla coronaria (L.) is found growing wild in northern Tunisia, while Sulla carnosa (Desf.)(Choi and Ohashi, 2003) grows wild in southern Tunisia.The first objective of this study was to characterise growth, Na + and K + levels, nutritional parameters and transport and pigments in collections of these two Sulla species when treated with several NaCl concentrations (0, 100 and 200 mM).Our second objective was to describe genetic diversity between these two accessions using ISSR markers and to associate these molecular marker patterns with salinity-dependent and salinity-independent physiological data.

Plant growth and salinity treatment
Seeds of S. carnosa and S. coronaria were collected from natural populations prospected throughout Tunisia.Seeds were scarified, then soaked in water for 24 h and put in pots containing a 1:2 (v:v) mixture of sand and peat.The pots were placed in a greenhouse and irrigated first with distilled water for 3 days, then for 47 days with a modified Long Ashton nutritive solution (Hewitt, 1966) diluted to 1/5th strength so that the final solution was composed of 1.5 mM MgSO4, 1.6 mM K2SO4, 0.3 mM K2HPO4, 0.3 mM KH2PO4, 3.5 mM Ca(NO3)2, 2.0 mM NH4NO3, 0.04 µM MnSO4, 0.5 µM ZnSO4, 0.05 µM H3BO3, 0.02 µM MoO3 and 3 µM FeEDTA.Plants were harvested at day 50; thereafter, the remaining plants were irrigated for an additional 27 days with the same nutritive solution supplemented with NaCl (0, 100 or 200 mM) and then harvested again.For molecular analyses, seeds were scarified and then soaked in water for 24 h, transferred to Petri dishes and incubated in a greenhouse for 4 days at 25°C and a 12 h day length until root and cotyledon emergence.

Growth and elemental analyses
Fresh and dry matter of leaves and roots were determined.Roots were collected by submerging the whole root system three times in 120 ml distilled water to remove growth substrate from the roots.Fresh and dry matters were determined for leaves and roots.Dried roots and shoots of each plant were then separately digested in 0.1 N HNO3.The concentrations of K + and Na + in the clear digests were measured by flame photometry (Jenway PFP7 butane air flame).The rates of ion net uptake (Jupt) and transport (Jtrp) were calculated according to Pitman (1988) as: ∆I tot Ln (Wr 2 /Wr 1 ) J upt = (t 2 -t 1 ) ∆ W Where, ∆ stands for the difference between times t2 (day and t1 (days); Itot is the average amount of ion per plant (mmol) calculated from ion concentrations (mmol g-1 DW) and plant dry weights (g) and Wr is the mean root dry weight (g).
∆I s Ln (Wr 2 /Wr 1 ) J trp = (t 2 -t 1 ) ∆ W Where, ∆ stands for the difference (in days) between time t2 (day 77) and time t1 (day 50); IS is the average amount of measured ion per shoot (mmol) calculated from ion concentrations (mmol g -1 DW) and shoot dry weights (g) and Wr is the mean root dry weight (g).
Growth and ion data were presented as the means of eight plants for each treatment (one of three salinity conditions for each of two plant species).Significant differences between treatments were determined by multi-variate analysis of variance (ANOVA) and means were compared and separated with a LSD test (Statistica  , StatSoft France)) at p≤0.05.

DNA extraction and PCR amplification
DNA was isolated from frozen fresh seedlings according to the procedure described by Dellaporta et al. (1983) with minor modifications adapted to mini-extraction.PCR amplifications were conducted using a Crocodile III thermocycler (QBIOgène, France) and three ISSR primers (Table 1).Each amplification was performed in a total volume of 25 µl and contained: 30 ng of total cellular DNA, 200 mM of each dNTPs (DNA polymerization mix, Pharmacia, France), 1.5 mM MgCl2, 60 pg of primers, 2.5 µl of 10x Taq DNA polymerase buffer and 1.5 U of Taq DNA polymerase (QBIOgène, France).Amplification conditions included an initial step of 5 min at 94°C, followed by 35 cycles (30 s at 94°C, 90 s at 45 to 60°C depending on the primer (Table 1) and 90 s at 72°C) and a final 5 min step at 72°C.ISSR PCR products were separated on 0.8% agarose gels and the DNA fragments were visualized by comparing ethidium bromide stained patterns and intensities with a DNA size marker of known concentration.
Each ISSR fragment (band) was scored as 1 when present and 0 when absent.Bands showing unambiguous polymorphism between salinity treatments and/or S. carnosa and S. coronaria were entered into a binary data matrix.Principal components analysis (PCA) was conducted on the ISSR patterns using XLSAS software (Statistical Analysis System, version 2009.3) (SAS, 2009).A Pearson correlation test and a Mantel test (Mantel, 1967) were used to test the statistical relationship between the presence or absence of specific ISSR amplicon patterns and the elements of the two matrices generated by the genetic, growth and physiological traits.

Effect of salt on growth and water content
The effect of NaCl (0, 100 and 200 mM) on the growth of the two Sulla species collected from different ecological locales is presented on Figure 1.After 27 days of treatment, 100 mM NaCl decreased plant dry matter equally in S. carnosa and S. coronaria.Differences were observed when plants were treated with 200 mM NaCl.Indeed, 200 mM NaCl decreased whole plant dry matter to < 20 and 40% of the untreated plant levels for S. carnosa and S. coronaria, respectively.Water content was much more variable in S. carnosa than in S. Kaddour et al. 14357 coronaria and masked any differences in shoot water content between the two Sulla accessions (Figure 2).Concerning the roots, a decrease in water content was observed only with 200 mM NaCl and was similar for both accessions (Figure 2).
Effect of salt on content, transport and selectivity of K + and Na + Since salinity can affect the rates of Na + and K + ion uptake and transport, the rates of ion uptake and transport were calculated for the two Sulla species by dividing the mean ion absorption per plant and per shoot over the course of the NaCl treatment by the mean root dry weight.Net K + uptake was identical for both accessions in the absence of NaCl treatment, although, K + transport was slightly higher for S. carnosa without salt (Figure 3).However, 100 mM NaCl decreased the net K + uptake and transport to a greater degree in S. carnosa when compared with S. coronaria.Both K + parameters showed even further decreases with exposure to 200 mM NaCl, but seemed to be totally inhibited by NaCl in S. carnosa, reaching negative values that suggested a loss of K + in this accession.K + accumulation was also significantly more reduced in both shoots and roots by NaCl treatments in S. carnosa as compared to S. coronaria, although, S. coronaria roots accumulated less of this ion overall in the absence of salt than S. carnosa (Figure 3).Na + uptake was similar for S. carnosa and S. coronaria when treated with 200 mM NaCl, although, S. carnosa accumulated slightly higher amounts of Na + at 100 mM NaCl (Figure 4).Nonetheless, differences between the two accessions were observed for Na + accumulation, especially in the shoots.After 27 days of NaCl treatment, S. carnosa accumulated more Na + in shoots than S. coronaria (Figure 4).Indeed, S. carnosa accumulated 4.5 and 6.3 mmol g -1 DM with 100 and 200 mM NaCl, respectively whereas S. coronaria accumulated only 3.2 and 3.7 mmol g -1 DM, respectively when exposed to similar levels of salt.In roots, differences between the two accessions were less pronounced than those observed for shoots (Figure 4).K + selectivity is estimated by the ratio between (K/K+Na) in the shoots or the roots and (K/K+Na) in the culture medium, and is a measure of the plant's ability to discriminate between these two ions, particularly when they are being absorbed by the roots.S. coronaria maintained a higher K + selectivity in the presence of 100 and 200 mM, NaCl when compared with S. carnosa for both roots and shoots (Figure 5).

Effect of NaCl treatment on chlorophyll and carotenoid content
Together with the impact of salt on growth, the photo-  system of a plant is usually compromised under saline conditions unless sufficient anti-oxidants are present.S. carnosa and S. coronaria showed equivalent levels of chlorophyll a, chlorophyll b, total chlorophyll and carotenoids (one type of anti-oxidant) in the absence of salt and when treated with 10 mM NaCl (Figure 6).S.
carnosa leaves showed a decrease in chlorophyll a, chlorophyll b, total chlorophyll and carotenoids when exposed to 200 mM NaCl, while these parameters were unaffected in S. coronaria leaves by the high salt concentration (Figure 6).The decrease in S. carnosa was estimated as 63% for chlorophyll a, 68% for chlorophyll Data is expressed as means ± confidence intervals (n=8, p ≤ 0.05).Statistically significant differences of the means are indicated by different letters (K + uptake), prime (') letters (K + transport), italicized letters (K content in shoots) or italicized prime letters (K + content in roots).
b, 64% for total chlorophyll and 46% for carotenoids when compared the non saline condition.

DNA profiling of S. carnosa and S. coronaria
Since physiological measurements indicated substantial differences in their responses to Salinity, DNA profiling was conducted on S. carnosa and S. coronaria to determine the degree to which these species were related.Three ISSR primers [(AG) 10 G, (AG) 10 T and (TC) 10 C], were used to analyse DNA polymorphism between S. carnosa and S. coronaria.Gel electrophoresis patterns of were used to analyse DNA polymorphism between S. carnosa and S. coronaria.Gel electrophoresis patterns of amplicons obtained using the three primers are illustrated in Figure 7 for 8 to 10 individual plants per Sulla species.For S. carnosa, a total of 64 ISSR loci were observed (only clear bands were scored), and all of them were polymorphic between the two accessions.For S. coronaria, 63 ISSR loci were scored, while for S. carnosa, all of these were polymorphic between the accessions (Figure 7).

Principal component analysis
Genetic variability between S. carnosa and S. coronaria was also studied by applying principal component analysis (PCA) applied to a data matrix developed for molecular marker data determined for the two accessions.axis 18.7%, 2nd axis 9.6% and 3rd axis 8.65%.Graphic representation of dispersion generated by the first two axes (28.3% of total variability) revealed strong genetic diversity between S. carnosa and S. coronaria (Figure 8).In this representation, the presence of 9 specific amplicon bands (positive molecular markers) was attributed to S. coronaria, while the presence of negative molecular markers denoted S. carnosa (Table 2).In particular, positive markers C2100, C2600 and C3900 from axis 1 were attributed to S. coronaria and the negative markers B860, A1000 and A1175 were attributed to S. carnosa.

Correlation analyses
Physiological and molecular data for both Sulla accessions were organized into a matrix and correlation coefficients determined using a Pearson correlation test.
The correlation values between the different salt treatment (0, 100 and 200 mM), showed raised values varying between -0.918 and 1 in the roots and between -0.961 and 1 in the shoots (data not shown), therefore, molecular data showed significant correlation with physiological responses in both accessions.The obtained matrix showed that correlation coefficients increased with increasing salt concentration and varied for both shoots and roots, with positive values varying from 0.7 to 0.9 and negative values varying from -0.6 to 0.9.Two molecular markers B340 and B860 were identified in roots and shoots after salt treatment using these correlations.The presence of these two markers were positively correlated to decreased Na + content and negatively correlated to increased DM, increased water content and increased K + content.The significance of these molecular markers to the overall physiological response of the two Sulla species was explored by a Mantel test.Statistical significance between physiological and molecular data were obtained for both shoots (r= 0.213, 0.003 < p < 0.05) and roots (r= 0.244, 0.002 < p < 0.05) (Figure 9), although, these overall correlations were relatively weak.

DISCUSSION
Salinity is one of the major stresses that severely limit crop production, especially in arid and semi-arid regions (Shannon, 1998).Sulla (Sulla and Hedysarum species) are leguminous plants with potential to be used as soil reclamation ground cover and livestock forage on arid lands.Their nitrogen fixing capacity reduces fertilizer inputs, and their moderate anthocyanin and proanthocyanidin content (Skadhauge et al., 1997) could protect against oxidative stress and impact positively on ruminant nutrition (Marles et al., 2003;Li et al., 2009).Two species of Sulla discovered growing in widely divergent ecological areas of Tunisia were investigated for their ability to withstand saline conditions The addition of NaCl 200 mM decreased plant growth more in S. carnosa than in S. coronaria.Reduction in plant growth can be related to excessive accumulation of Na + (Fernandez-Ballester et al., 1998).In shoots, S. carnosa accumulated more Na + than S. coronaria.When comparing NaCl sensitivity among barley genotypes, Flowers and Hajibagheri (2001) have found that the more sensitive variety of barley triumph accumulated more Na + in the shoot than the tolerant variety Gerbel.Kaddour et al. 14363 To maintain cytosolic Na + homeostasis, excess ions must be removed from the cytosol either to the vacuole (intra-cellular compartmentalization), or be exported from the cell (Tyerman and Skerrett, 1999).In this study, Na + transport and uptake were compared between the two Sulla species and the results showed a higher Na + transport in S. carnosa when compared with S. coronaria, at NaCl 100 and 200 mM.Na + transport was also compared between the relatively salt tolerant landrace line 149 and the salt sensitive cultivar Tamaroi.It was found that the major differences in Na + transport between the two varieties of durum wheat (Triticum turgidum) were; the rate of transfer from the root to the shoot (xylem loading), which was much lower in the salt tolerant genotype and the capacity of the leaf sheath to extract and sequester Na + as it entered the leaf (Davenport et al., 2005).Salt concentrations of 100 and 200 mM decreased K + accumulation more in S. Carnosa.The decrease in K + accumulation could be related to a more inhibited K + transport in this accession.Potassium plays a number of important roles in plant growth and development.
Molecular approaches have greatly advanced our understanding of K + transport in plants and a large number of genes encoding K + transport systems have been identified.In A. thaliana for example, at least 35 genes are thought to encode various K + channels or transporters (Qi and Spalding, 2004).The maintain of a high cytosolic K + /Na + ratio is considered as a criteria of salt tolerance in many species (Cuin et al., 2003;Gorham et al.,1990;Rubio et al., 1999).Differences in Na + and K + accumulation between the two Sulla species, could explain the better K/Na ratio in S. coronaria.NaCl addition decreased plant salt tolerance, and is likely to be conferred by a large number of adaptive mechanisms.Among the most important of these is the ability of the plant to maintain a high K/Na ratio in the cytoplasm as well as to keep cytosolic Na + content below some crucial value (Greenway and Munns, 1980;Maathuis and Amtmann, 1999;Tyerman and Skerrett, 1999).In A. thaliana, salt tolerance of the NOK2 ecotype was related to the maintaining of a higher K/Na ratio when compared with Col ecotype (Kaddour et al., 2009).
Salinity modifies the level of secondary metabolites, such as caraotenoids (carotenes and xanthophylls), having antioxidant properties and acting as accessory light harvesting pigments (de Pascale et al., 2001).Our results show a significant decrease in carotenoids and chlorophyll content in S. carnosa, while the level of these two metabolites remained unaffected by NaCl 100 and 200 mM in S. coronaria.The decrease in chlorophyll content is considered as an indicator of salt stress (Singh and Dubey, 1995).In mungbean, steady levels of CAR are strongly correlated to its salt tolerance (Wahid et al., 2004).
The ISSR markers measured diversity among accessions at DNA level, thus no interaction with environment was expected (Terzopoulos and Bebeli, 2008).All the generated ISSR loci were polymorphic between the two accessions confirming that ISSR technique is very efficient to examine genetic diversity between these two accessions.
S. carnosa showed a large projection on the axis 1 and 2, in relation to its big diversity that could be explained by its large geographic distribution.Physiological characterization showed a sensitivity of S. carnosa to salt stress.Therefore, the important DNA polymorphism put in evidence in this accession seemed to be related more to its water stress resistance rather to its salt tolerance.
However, molecular markers found in S. coronaria, the less salt sensitive accession would be more in relation to its salt tolerance.
The implication of the molecular markers in the physiological response of Sulla, were explored by Mantel test.The results show that the statistically significant correlation between physiological and molecular data has revealed two molecular markers (B340 and B860) that are respectively identified in the roots and shoots after salt treatment.These two markers were positively correlated to salt response of Sulla accessions.
In conclusion, Sulla coronaria showed better salt tolerance on the basis of growth and K/Na selectivity that makes the local accession good candidates for a future breeding programme.

Figure 1 .Figure 2 .
Figure 1.Effect of NaCl treatment on the growth of S. carnosa and S. coronaria shoots and roots.NaCl treatments (0, 100 and 200 mM) were applied to 50 day-old individual plants and lasted for 27 days.Data is expressed as means ± confidence intervals (n=8, p ≤ 0.05).Statistically significant differences of the means are indicated by different letters for whole plant measurements (but not for individual shoot and root tissues).

Figure 3 .
Figure3.Effect of NaCl treatment on K + content, uptake and transport into roots and shoots of S. carnosa and S. coronaria.NaCl treatments (0, 100 and 200 mM) were applied to a 50 day-old individual plants and lasted for 27 days.Data is expressed as means ± confidence intervals (n=8, p ≤ 0.05).Statistically significant differences of the means are indicated by different letters (K + uptake), prime (') letters (K + transport), italicized letters (K content in shoots) or italicized prime letters (K + content in roots).

Figure 4 .Figure 5 .Figure 6 .Figure 7 .
Figure 4. of NaCl treatment on Na + content, uptake and transport into roots and shoots of S. carnosa and S. coronaria.NaCl treatments (0, 100 and 200 mM) were applied a 50 day-old individual plants and for 27 days.Data is expressed as means ± confidence intervals (n=8, p ≤ Statistically significant differences of the means are indicated by different letters (Na + uptake), prime (') letters (Na + transport), italicized letters (Na + content in shoots) or italicized prime letters (Na content in roots).

Figure 9 .
Figure9.Correlation analysis between physiological data for shoots (left) and roots (right) and ISSRs markers according to a Mantel significance test.Statistically significant correlation between physiological and molecular data were obtained for shoots (r= 0.213, 0.003 < p < 0.05) and roots (r= 0.244, 0.002 < p < 0.05).

Table 1 .
Percentage of polymorphic bands.

Table 2 .
Principal component analysis of Sulla ISSR amplicon bands.
Figure 8. Principal component analysis plot illustrating the first two variability components (axes) for Sulla carnosa and Sulla coronaria ISSRs markers.S. carnosa: black symbols and S. coronaria: white symbols.