Potential determinants of salinity tolerance in rice (Oryza sativa L.) and modulation of tolerance by exogenous ascorbic acid application

Rice is a relatively salt-sensitive crop with the reproductive and seedling stages being the most sensitive. Two separate experiments were conducted to isolate potential determinants of salinity tolerance and to investigate the possibility of modulating salt tolerance by exogenous ascorbic acid (AsA) application. Rice plants were imposed to salinity (EC= 10.0 dS m-1) both at the seedling and reproductive phases of growth. Salinity at the seedling stage resulted a sharp decline in shoot and root growth related traits including leaf chlorophyll content, while hydrogen peroxide (H2O2) and malondialdehyde (MDA) levels increased. Plants experienced with salinity at the reproductive phases of growth showed a significant reduction in yield attributing traits while the tissue levels of H2O2 increased. Exogenous AsA application reversed the negative impact of salt stress, modulating the root and shoots growth and yield related traits and lowering H2O2 and MDA levels. FL-478 was identified as the most tolerant genotype at the seedling stage, with Binadhan-10 being the most tolerant at the reproductive stage. Grain yield panicle-1 significantly and positively corrected with number of filled grains panicle-1, panicle length, plant height, and spikelet fertility, and negatively correlated with H2O2 levels. Stress tolerance indices clearly separated the tolerant and susceptible genotypes. A principal component analysis revealed that the first two components explained 87% of the total variation among the genotypes. Breeding efforts could therefore to undertake for developing salinity tolerance by manipulating endogenous AsA content in rice.


INTRODUCTION
Saline soils are one of the most severe constraints to crop productivity worldwide and thus are a major concern for global food security [1].It has been projected that more than 20% of the world's arable land and 50% of irrigated areas, which include around 30% of rice growing areas, already suffer from salinity problems [2]. To feed an increasing global population, it will be essential to utilize these saline soils either by using reclamation to reduce salinity or by growing salt tolerant crop plant [3]. Conventional plant breeding to increase crop yields in saline environments is often slow, due to our poor understanding of the molecular and genetic mechanisms of salt stress tolerance as well as lack of suitable phenotyping and genotyping techniques [4].
Hence, there is a pressing need to improve our understanding of the complex mechanisms associated with salt tolerance, as well as to develop appropriate phenotyping and genotyping techniques to be used for the development of modern crop varieties that are more resilient to salt stress. An understanding of salinity tolerance mechanisms at the various phases of plant development and identification of potential traits associated with salt stress determinants by using genotypes with variable degrees of salt stress tolerance will enable us to develop robust salt tolerant crop varieties or management techniques for sustainable crop production [5].
food of most people in Bangladesh and it is the world's second most important grain. In Bangladesh, rice occupies about 70% of the total cropped area, about 13.9 million hectares. Of the 2.85 million hectares of coastal arable land about 1.056 million hectares are affected by various salinity levels and crop production in those areas is very limited [6]. Importantly, it has also been predicted that increases in soil salinity may lead to a decline in rice yield by 15.6% by 2050 [7]. So, there is a critical need for developing rice varieties that can withstand high levels of salt stress and maintain satisfactory yields under both saline and non-saline conditions.
Rice is relatively tolerant to stress during seed germination, active tillering, and at maturity but is very sensitive at the early seedling stage and at the reproductive stage [2,8]. Very poor correlation exists between tolerance at the reproductive and seedling stages, suggesting that these two sensitive stages are independent of each other and are controlled by dissimilar sets of genes [9]. Despite much research having been conducted for the salt tolerance at the seedling stage very little attention has been paid to the reproductive stage, although the reproductive stage is most decisive as it finally governs grain yield [10]. This is mainly because of a lack of reliable reproductive-stage-specific phenotyping protocols [2].
Salinity imposes ionic and osmotic stresses on plants [11] and increased levels of reactive oxygen species (ROS), such as hydrogen peroxide (H 2 O 2 ), superoxide (O 2 .-) and hydroxyl radical, and this ultimately leads to oxidative stress [12][13][14]. Consequently, metabolic dysfunction and damage to cellular structures, inhibition of photosynthesis contribute to growth perturbances, reduced fertility, and premature senescence in plants [15]. The most injurious effect of salinity, at the reproductive stage, is on panicle initiation, pollen viability, spikelet formation, pollen germination and fertilization, and significant effects have been observed on panicle weight, panicle length, primary branches panicle -1 , number of filled grains panicle -1 , number of unfilled grains panicle -1 , total grain number panicle -1 , total grain weight panicle -1 , 1000-seed weight and total grains plant -1 [16][17][18]. Though pollen and spikelet fertility and pollen germination are the most important determinates of rice yields under salt stress [19,20], there are no comprehensive reports concerning the relationship between ROS metabolism and pollen or spikelet fertility in response to salinity, although high Na + concentrations are known to be associated with pollen sterility [21]. Importantly, drought induced spikelet sterility was reported to be associated with an abrupt increase in ROS and/ or inefficient antioxidant defenses [22]. Enhanced antioxidant defense is one of the mechanisms plants use to adapt to adverse environments, including salt stress [13][14]23]. Changes in the amount and the activities of antioxidant enzymes in response to salinity were found to differ between salt sensitive and tolerant cultivars of various crop plants [23,24]. While several studies have been conducted to decipher the mechanisms of salt tolerance at the seedling stage, little emphasis has been given to the biochemical mechanisms of salinity tolerance during panicle initiation and flowering. Therefore, assessment of popular salt tolerant and high yielding rice varieties, tolerant breeding lines and sensitive varieties, for spikelet fertility in relation to ROS metabolism could open up new possibilities to breed rice varieties for higher salt tolerance.
Ascorbic acid (AsA) is a highly water-soluble antioxidant molecules and plays a vital role in plant defense, including ROS detoxification through the ascorbate-glutathione pathway and through cellular signalling that triggers adaptive responses [25]. Optimum levels of glutathione and AsA were found to improve the overall productivity of plants through the modulation of osmoregulation, water use efficiency, photosynthetic performance and plant water and nutrient utilization efficiency [25]. Under stressful conditions higher rates of AsA degradation have been observed and without increased rates biosynthesis this can cause an imbalance in cell redox homeostasis [26][27]. Exogenous application of AsA was found to improve salt tolerance in various crop plants through protection of lipids and proteins from oxidative damage, particularly at seedling stage [28][29][30]. Transgenic plants over-expressing AsA biosynthetic pathway genes have been shown to possess higher abiotic stress tolerance, including salinity tolerance, by improving ROS and methylglyoxal(MG) detoxification [31][32][33]; however, many aspects of the role of exogenous AsA in modulating salt stress tolerance particularly at reproductive phase is remain unknown. In rice, research related to spikelet fertility/sterility in relation to salt-induced ROS accumulation and AsA metabolism is unknown. Considering the above, the present research was conducted to identify potential morphological and biochemical determinants and stress tolerance indices in rice at the seedling and reproductive stages, using contrasting genotypes having difference in salinity tolerance, using an appropriate phenotyping protocol. The potential of exogenous AsA-mediated salt stress tolerance was also investigated as an addition to future breeding strategies.

Seedlings Growth Conditions and Salt Stress Treatments
Two separate experiments were conducted with four replications and four treatments (viz., control, 2mM AsA, salinity (EC= 10.0 dSm -1 ) and salinity +2mM AsA) under a completely randomized design (CRD). In experiment I, rice seedlings were cultured in professional peter solution under hydroponic condition. After 13 days of seedling growth, two groups of seedlings of each genotype were pre-treated with 2mM AsA in peter solution for 24 h. After 14 days of seedling growth, AsA pre-treated and untreated seedlings were then imposed to salinity (10 dS m -1 NaCl solution) in hydroponic peter solution for 96 hours. The other group of AsA pre-treated seedlings was also grown under control condition.
In experiment II, rice plants were grown in perforated pots filled with field soil and at the end of the rice growth stage 4 (young panicle about to emerge from flag leaf) two groups of rice plants of each genotype were pruned, leaving the penultimate leaf and flag leaf, and then subjected to 10 dS m -1 salt stress in hydroponic tank and allowed to grow up to the ripening stage [2]. One group of salt-stressed rice plants and one group of control rice plants were sprayed with a 2mM AsA solution containing 0.001% Tween-20, using a hand sprayer, after 2 days and 4 days of salinity treatment. The same amount of distilled water was sprayed onto control plants.

Data on Morphological Traits Recorded at the Seedling and Reproductive Stage
In experiment I, data on shoot length, root length, fresh root weight, and fresh shoot weight were calculated from ten seedlings per genotype for each replication. Data on dry root weight and dry shoot weight were estimated after oven-drying of the samples at 60 o C for 3 days.
In experiment II, data on yield and yield attributing traits (days to maturity, panicle length, plant height, number of unfilled grains panicle -1 , number of filled grains panicle -1 , spikelet fertility (%), 100-seed weight, grain yield panicle -1 ) were recorded from ten plants replication -1 for each genotype, after harvesting. Data on H 2 O 2 was measured from flag leaf tissues after 8 days of salt stress.

Determination of Chlorophyll C ontent
Chlorophyll content was determined from leaf tissues (0.5 g) of the seedlings by soaking in 80% acetone as described by Sohag et al. [34]. The absorbance of the acetone extracts was measured at 645 and 663 using a UV-VIS spectrophotometer (Shimadzu, UV-1201, Japan). The total chlorophyll content was expressed as mg g -1 FW.

Determination of Hydrogen Peroxide (H 2 O 2 )
Hydrogen peroxide from leaf tissues at the seedling stage and from flag leaf tissues at the reproductive stage was measured following the method of Velikova et al. [35] and the H 2 O 2 content was calculated by utilizing 0.28 μM -1 cm -1 extinction coefficient [36].

Determination of Malondialdehyde (MDA)
MDA was measured from leaf tissues (at seedling stage) following the standard method as described by Heath and packer [37] and the MDA content was determined utilizing 155 mM -1 cm -1 extinction coefficient and expressed as nmolg -1 FW [38].

Statistical Analysis
Data analysis was carried out using the Minitab

Effect of AsA, Salt Stress and Salt+2mM AsA Treatments on Rice Genotypes at the Seedling Stage
The results of analysis of variance for all the characters (viz., root length, shoot length, fresh root weight, fresh shoot weight, dry root weight, dry shoot weight, total chlorophyll, H 2 O 2 , MDA) showed highly significant (P≤0.001) variations due to genotypes as well as treatments (Supp. Table 1). Root length, shoot length, fresh shoot weight, dry root weight, dry shoot weight, total chlorophyll, H 2 O 2 , MDA were found to have significant (P ≤0.001) G × T interactions, whereas shoot fresh weight showed no significant G × T interaction (Supp. Table 1).

Root Length
The greatest root length under control conditions was found for BRRI dhan28 (12.23 cm), whereas the lowest was found for Binadhan-8 (7.32 cm) ( Table 1). Root length showed a significant decrease under salt stress in comparison with control for all the genotypes studied. The greatest reduction was observed in salt susceptible Binadhan-6 (15.51%), whereas the least reduction was found for Binadhan-10 (0.19%) (

Fresh Shoot Weight
The greatest fresh shoot weight was found for FL-478 (460.00 mg) and the lowest for Binadhan-8 (220.00 mg) ( Table 1) under control conditions. In response to salinity, shoot fresh weight showed a significant decrease in all of the genotypes, with the greatest reduction for Binadhan-6 (54.74%) and the least reduction for Binadhan-10 (22.13%), as compared to controls. AsA pre-treatment was found to reduce the decrease of fresh root weight under the condition of salinity. The highest decrease was found in BRRI dhan28 (33.34%) followed by Binadhan-6, BRRI dhan67, Binadhan-10, Binadhan-8, BRRI dhan78 and FL-478 (23.26, 20.45, 13.64, 11.76, 11.11 and 7.69%, respectively) as compared to the seedlings treated with salt stress only (Table 1).

Dry Root Weight
The maximum dry root weight was observed in FL-478 (42 mg) and minimum in Binadhan-8 (12 mg) under control treatments

Hydrogen Peroxide
The

Malondialdehyde (MDA)
The highest level of MDA was found in in Binadhan-6 (33.74 nmol g -1 FW) and the lowest in BRRI dhan78 (23.06 nmol g -1 FW) under control conditions (

Effect of AsA, Salt Stress and salt+2mM AsA Treatments on Rice Genotypes at the Reproductive Stage
The result of the analysis of variance for all the characters showed highly significant (P ≤0.001) variation due to genotypes and treatments (Supp. Table 2). Days to maturity, numbers of filled grains panicle -1 , number of unfilled grains panicle -1 , spikelet fertility, 100-seed weight, grain yield panicle -1 , H 2 O 2 also showed significant (P ≤0.001) G × T interactions, whereas plant height and panicle length showed no significant G × T interaction (Supp. Table 2).

Days to Maturity
The maximum number of days to maturity under control conditions was found for BRRI dhan78 (164), whereas the least was found for BRRI dhan67 (148) ( Table 2). A significant decrease in days to maturity was observed under salinity stress in all the genotypes studied. The greatest reduction was observed in salt susceptible Binadhan-6 (11.40%), whereas the least reduction was found for BRRI dhan78 (8.10%) (  (Table 2). AsA treated non-stressed genotypes showed no significant differences in panicle length in comparison with controls.

Number of Filled Grains Panicle -1
The highest number of filled grains panicle -1 was found for Binadhan-6 (109.37) under control conditions and the lowest was for FL-478 (62.74) (

Number of Unfilled Grains Panicle -1
The maximum number of unfilled grains panicle -1 was observed for BRRI dhan78 (31.82) and minimum for BRRI dhan28 (14.84) under control treatments (

Estimation of Correlation Co-efficient Among Nine Characters of Rice Genotypes Under Control and Salt Stress Conditions at the Reproductive Stage
Days to maturity showed a significant positive correction with plant height and panicle length under both control and salt stress conditions, but showed significant positive correlation with number of filled and unfilled grains panicle-1 and negative correlation with 100-seed weight and H 2 O 2 content under control conditions. Days to maturity showed a significant positive correction with grain yield panicle -1 under stress conditions (Table 3). Plant height showed a significant positive correlation with panicle length and number of filled grains panicle -1 and grain yield panicle -1 under both control and salt stress conditions, but a significant negative correlation with 100-seed weight under control conditions. Plant height showed a significant positive correction with grain yield panicle -1 and negative correlation with H 2 O 2 ( Table 3). Panicle length showed a significant positive correction with number of filled grains panicle -1 under both control and salt stress conditions, but a significant negative correlation with 100seed weight under control conditions. Under salt stress conditions, panicle length showed a significant positive correlation with spikelet fertility and grain yield panicle -1 , but showed a significant negative correlation with 100-seed weight and H 2 O 2 content. Number of filled grains panicle -1 showed a significant positive correlation with spikelet fertility and grain yield panicle -1 under both control and stress conditions, but showed significant negative correlation with 100-seed weight under control conditions. Under salt stress conditions number of filled grains panicle -1 showed a significant negative correlation with number unfilled grains panicle -1 and H 2 O 2 content. Number of unfilled grains panicle -1 showed a significant negative correlation with spikelet fertility under both control and stress conditions, but showed significant negative correlation with grain yield plant -1 and H 2 O 2 under stress condition. Spikelet fertility showed a significant positive correlation with grain yield panicle -1 and negative correlation with H 2 O 2 under salt stress conditions. 100-seed weight and grain yield panicle -1 showed a significant negative correlation with H 2 O 2 under salt stress conditions (Table 3).

Principal Components (PCs) for Nine Morphological and Biochemical Traits in Seven Rice Genotypes from PCA
The first two principal components PC1 and PC2 explained 66.5% and 20.5% of total variation, respectively (Table 4). Because PC1 collectively explained more than half (66.5%) of the variation and contributed more to the separation of genotypes into different categories, they were used to classify the 84 groups of 7 rice genotypes into four major groups including highly salt sensitive, moderately salt sensitive, moderately salt tolerant and highly   salt tolerant. From the biplot, it was found that the PC1 scores of FL-478 under the salt stress treatment completely separated from those of Binadhan-6 under control and AsA treatments. The variation between FL-478 under control conditions and Binadhan-6 under control and AsA treatments were due to a higher negative coefficient of the traits: number of unfilled grains panicle -1 (UFG/P) and H 2 O 2 compared to the positive coefficients of the traits of number of filled grains panicle -1 , grain yield panicle -1 and spikelet fertility (Figure 1). Similarly, PC2 scores of BRRI dhan78 under Salt+ 2mM AsA treatment completely separated from those of FL-478 control and AsA treatments, due to higher positive coefficients of the traits of plant height, panicle length, number of unfilled grains panicle -1 , H 2 O 2 , days to maturity compared to the negative coefficients of the traits of 100-seed weight.

Ranking based on Morphological and Biochemical Traits of Rice Genotypes at the Seedling and Reproductive Stages
Considering all the traits at the seedling stage under control conditions, FL-478 was in first position followed by

DISCUSSION
Rice is currently registered as the most salt-sensitive cereal crop with a threshold of 3dSm -1 for most cultivated varieties [43]. Salinity caused a substantial reduction in plant development and growth as compared to respective control plants during all growth stages and it prevented plants from fully expressing their full genetic potential [44]. Salt stress triggers a reduction in intra-cellular water potential and water availability and so roots fail to absorb sufficient water and nutrients for adequate plant growth [45]. The current study showed that the imposition of salt stress significantly reduced the shoot growth and root characteristics and as well as leaf chlorophyll content ( Table 1). The reduction of shoot and root growth and chlorophyll content was greatest in the sensitive cultivars compared to tolerant genotypes. Salinity reduced growth of plants obviously due to the negative consequence of salt stress that restricts cell division [46] and arrests plant growth at least in part due to the initiation of oxidative stress [47]. Additionally, the reduction in development and growth of salt-stressed seedlings could be due to the negative effects of the high osmotic potential of the nutrient solution that lowered uptake of water and nutrients [48]. The reduction in morphological parameters and chlorophyll content due to salinity was also found by other researchers [36,[49][50][51][52]. The findings of the present experiment also show that the salt treatment led to increased H 2 O 2 and MDA contents in all of the genotypes tested, but that the accumulations of H 2 O 2 and MDA were lower in salt-tolerant genotypes compared to salt-sensitive genotypes ( Table 1). The lower accumulation of MDA and H 2 O 2 in the salttolerant genotypes implies greater protection against oxidative damage by better regulating mechanism to ROS formation to perform their signaling function [53,54] and therefore, these genotypes displayed more salinity tolerance [23,55,56]. In contrast, the higher accumulation of H 2 O 2 and MDA contents in salt-sensitive genotypes was probably due to higher rates of ROS production as well as inactivation of antioxidant enzymes [57], leading to oxidative stress and membrane lipid damage [58,59]. Generation of oxidative stress in response to short term salt stress in rice was also reported by others [13,60]. Importantly, an exogenous AsA pre-treatment resulted in greater root and shoot length as well as chlorophyll content for plants of all the genotypes tested, compared to plants treated with the salt stress only (Figure 2). Similar results were also indicated by other researchers [49,[61][62][63][64]. Exogenous AsA led to a reduction in ROS and MDA content, however the greatest reduction was noted in the sensitive genotypes in comparison with tolerant genotypes that could mean that the tolerant genotypes synthesize more AsA in comparison with sensitive genotype or produces lower ROS levels. Therefore, the higher growth rates of AsA pre-treated seedlings might due to efficient ROS detoxification and /or lower ROS synthesis and accumulation, and also better ion homeostasis, particularly the maintenance of low Na + /K + ions through the mechanisms like salt exclusion, ion partitioning and compartmentation of Na + into shoots [61] and the proper signaling function of AsA and ROS for up-regulation of stress responsive genes [65][66][67] (Figure 3). Significant variations in days to maturity due to treatments within a particular genotype were found because of direct salt inclusion from root zone to panicle. Salt stress forced the plants to mature earlier, but tolerant genotypes showed a similar number of days to maturity when grown under control or salt stress conditions. The yield contributing traits like number of filled grains panicle -1 , panicle length, plant height, spikelet fertility %, 100-seed weight, and grain panicle -1 and yield were significantly reduced by the imposition of salt stress whereas the number of unfilled grains increased ( Table 2). The reduction of yield attributing traits and yield due to salinity were also mentioned by other researchers [18,59,68,69]. Importantly, application of exogenous AsA was found to improve yield attributing traits and yields under salinity stress condition. The increase in yield and yield attributing traits due to AsA under salt stress were also found by other researchers [28]. Importantly the level of H 2 O 2 increased significantly however the highest increase noted in the genotype those were salt susceptible. An increase in the MDA content in flag leaves at the reproductive stage in response salt stress was also reported by Moradi and Ismail [70] and similar results were found by the others [49,61]. Application of exogenous AsA was also found to lower H 2 O 2 and MDA level under salt stress in other studies [67,71,72].
The phenotypic correction (Table 3) study among the yield attributing and yield traits reflects a significant positive correlation of grain yield panicle -1 with the other morphological traits studied, whereas yield panicle -1 showed significant negative correlation with the biochemical traits measured (ROS and MDA). A similar positive correlation with yield was also reported by others [73][74][75][76]. The increase panicle length, filled grains and spikelet fertility ensures increased grain numbers, which contributes to increased grain yield. Furthermore, positive correlation of 100-seed weight with yield indicates the importance of individual grain weight for increased yields. However, the number of unfilled grains panicle -1 showed a PCA provide an explanation and indication of the decisive component traits contributing to salt tolerance for the genotypes and treatments under study [77]. In this research Common negative effects of salt stress on plants are ionic imbalance and osmotic stress, which can trigger the accumulation of toxic compounds, e.g. reactive oxygen species (ROS), and depletion of antioxidants, e.g. AsA. These negatively affect plant growth and development, and consequently yield-contributing parameters and final yields. Application of exogenous AsA to plants under salt stress resulted in the maintenance of non-toxic levels of ROS, all of which contributed to the alleviation of salt-induced damage, leading to enhanced growth and development, and resulted higher yields experiment, PCA analysis disclosed that PC1 is negatively correlated with number of unfilled grains panicle -1 and H 2 O 2 ( Table 4). This is in contrast to grain yield panicle -1 , spikelet fertility, number of filled grains panicle -1 , days to maturity, and 100-seed weight, which are positively correlated with PC1. Therefore grain yield panicle -1 , spikelet fertility, number of filled grains panicle -1 , days to maturity, and 100-seed weight traits are positively associated with salinity tolerance in the present study. Under these hypothetical conditions, the H 2 O 2 level was negatively correlated with spikelet fertility and 100-seed weight.
The results suggest that plants with higher H 2 O 2 level contents have lower spikelet fertility and 100-seed weights under saline conditions. Genotypes showing the highest values for the positively correlated traits for PC1and PC2, were considered as highly saline tolerant genotypes and taking the place in the upper-right corner of the biplot. Genotypes with moderate values for PC1 and PC2, located in the lower right and upper left corner of the graph, were considered as moderately salt tolerant and moderately salt sensitive, respectively ( Figure 1). On the contrary, genotypes displaying the low values of positively correlated traits fall in the lower left portion of the biplot and were categorized as salt sensitive. Similar four major groups also categorized by Kakar et al. [78] using 74 rice genotypes. Under salt stress + AsA condition, only Binadhan-10, a salt tolerant rice genotype, is in the upper-right corner of the biplot, which has both positive effects for PC1 and PC2 and this genotype is therefore classified as highly saline tolerant genotype.
Stress tolerance indices such as SSI, TOL, STI and YSI values estimated from grain yield panicle -1 were found to be effective in separating the susceptible and tolerant genotypes (Table 5). Among the stress tolerance indicators, Krishnamurthy et al. [79] suggested that higher values of TOL and SSI represent relatively more sensitivity to stress, thus a reduced value of TOL and SSI for a given genotype indicates the higher stability of the genotype in stress and no stress environments. Selection based on these two criteria favours genotypes with high yields under stress conditions. On the other hand, higher values of STI and YSI represent relatively a more tolerant genotype under salt stress than genotypes with lower values [80] and lower grain yield stability in stress conditions [81]. From the stress tolerance indices, FL-478 was the most susceptible genotype whereas Binadhan-10 was the most tolerant at reproductive stage of the seven rice genotypes tested. These selection indices are therefore effective for separating salt susceptible and tolerant genotypes. Additionally, individual ranking of the genotypes considering all of the traits at two different phases of plant growth indicates that tolerance at the reproductive and seedling stages are not correlated. Similar results were also reported by others [68,82].
In conclusion, our studies clearly demonstrated that salinity at the seedlings stage and/or reproductive stages significantly impacted plant growth and development, and reduce yields and yield attributing traits. A clear genotypic difference in salt tolerance was observed with respect to developmental stage. Exogenous AsA application at the seedling and /or reproductive stages improved salinity tolerance in rice through the positive modulation of growth, yield and yield attributing traits (summarized in figure 3). The findings of our current study are useful for rice breeding programs and further research on molecular aspects of AsA-mediated salinity tolerance in rice is warranted, with the aim to enhance endogenous AsA levels using a genetic engineering approach. However, field trials with various concentrations of AsA and salinity levels will be needed to provide more definite information for management of salt stress in rice, as well as for the genetic manipulation of rice plants to enhance productivity.