Wheat Antioxidants Changes and Grain Yield Variation under Drought Stress

Antioxidants provide a defense line against adverse effects of free radicals released in plant cells under drought stress. Accumulation of antioxidants may link with higher grain yield in environmental stress conditions. The aim of this study was to evaluate antioxidants accumulation under drought stress and their relationship with grain yield variation. Reactions of 100 wheat landrace varieties and two commercial cultivars (Shiraz and Bezostaya) to drought stress were tested by two treatments’ including normally irrigation regime throughout the growth cycle and stopped irrigation from heading stage as drought stress in 2010-2011. Results indicated that drought stress triggers plant reactions via accumulation of proline as non-enzymatic and catalase and peroxidase as enzymatic-antioxidants. Correlation analysis showed that grain yield increases when antioxidants increase and proline was more responsive than other antioxidants in plants dealing with water deficit stress after heading. Genotypes had different reactions to drought stress and categorized in susceptible and tolerant groups. KC4144, KC4641, KC4779, KC3885, KC4529, KC4863, KC4907, KC4528, KC4511 and KC4840 had higher enzymatic activities. These genotypes that had highest grain yield in both irrigation and drought stress conditions can be used as parental lines in construction of mapping populations in order to locate QTLs responsible for drought tolerance. Therefore, categorizing studied genotypes as drought- tolerant and susceptible would help a better programming of breeding for higher grain yield and stability under low-income environments.


INTRODUCTION
Drought stress is a crucial factor limiting the survival, growth and distribution of plants in the world [1]. Water deficit leads to the perturbation of all or some of physiological and biochemical processes and consequently reduces plant growth and yield [2,3]. Antioxidants are known to increase under drought conditions. In some cases, synthesis of new iso-enzymes is induced to overcome the negative impact of progressive water stress [4,5]. The relation between drought stress and enzymatic antioxidant systems has been studied in some plant species [6]. It has been reported that glutathione reductase helps plants in dealing with desiccation or drought. Plants have evolved a series of non-enzymatic and enzymatic antioxidants to cope with drought stress and to avoid photo-oxidative damage, either by stress avoidance or stress tolerance [7]. Reactive oxygen species (ROS) such as superoxide anion (O• 2 -), H 2 O 2 , and hydroxyl ( . OH) are commonly generated and accumulated when abiotic stress occurs [8][9][10]. Excessive levels of ROS damage cellular structures and macromolecules, causing photoinhibition of the photosynthetic apparatus [11]. However, the accumulation of ROS activates multiple defense responses. The metabolism of ROS depends on several functions involved in deactivation of hydrogen peroxide, though its affinity to Н 2 О 2 is low [12].
Among the common responses of plants to abiotic stresses is the production of different types of organic solutes including small molecules such as proline [13][14][15][16]. Proline is one of the most common osmolytes. Other antioxidants are superoxide dismutase (SOD), peroxidase (POD), catalase (CAT) and ascorbate peroxidase (APX) [17]. SOD removes superoxide anion free radicals accompanying the formation of H 2 O 2 , which is then detoxified by CAT and POD. In the ascorbate-glutathione cycle, APX reduces H 2 O 2 using ascorbate as an electron donor [18]. Correlation of the increased content of intracellular proline and the plant ability is critical to survive under high salinity and water limited conditions [19,20].
Under drought conditions, proline synthesis may provide reserves of organic nitrogen that are mobilized during plant recovery after resumption of normal water supply. Plant resistance to water deficit stress varies depending on plant species and the variety of the same species [21,22]. Those so-called osmotic regulators or compatible osmolytes protect plants from stresses by cellular adjustment through the protection of membranes integrity and enzymes stability [23,24]. A number of studies using transgenic plants demonstrated that proline has a complex effect on stress responses, suggesting that proline is important in stress tolerance [25,26]. For instance, it has been found that proline concentration was higher in stress-tolerant wheat cultivars than in sensitive ones [27]. It has been reported that proline has the ability to act as molecular chaperone protecting protein integrity and preventing protein aggregation and stabilization [28]. It also protects the nitrate reductase under osmotic stress conditions [29]. Proline plays an important role in stabilizing membranes antioxidant [30] and regulating the cytosolic acidity [5]. Usually, enhanced anti-oxidative protection is related to higher drought resistance [6]. A loss of membrane integrity and oxidative damage to lipids were more pronounced in the sensitive wheat varieties challenging with drought under field condition.
Therefore, the aim of present study was to evaluate the antioxidants response and grain yield variations in wheat landrace varieties in relation to drought stress conditions. The other purpose was to categorizing wheat genotypes in terms of their reactions to water deficit conditions for their possible uses in cross breeding programs which is the primary step of locating genes responsible for drought tolerance.

Plant Material
In order to evaluate changes in enzymes activity under drought stress at heading stage, an experiment with 100 Iranian landrace varieties (collected from different parts of Iran by the Seed and Plant Improvement Institute, Karaj, Iran) and two commercial cultivars (Bezostaya and Shiraz) as control was carried out in the research farm station of Shiraz University in 2010 -2011 growing season. The experiment was set out as a split plot based on randomized complete block design with three replications. The soil texture was sandy clay with electric conductivity of 0.36 deci Siemens m -1 and pH 7.0. The two water regimes were well-watered or normal irrigation (FC) and drought stress (50% FC). Water regimes and landrace varieties were assigned to main and sub plots, respectively. Seeds were sown in two 3 m long rows. Drought stress regime (50% FC) started by stopping irrigation at the early heading stage. In drought stress plots, irrigation was done every 8 days to keep 50% FC. In drought stress plots, soil water content was continuously measured by sampling from soil to keep 50% FC as drought stress treatment. Number of irrigation practices and the amount of water supply were similar for pre-heading stages in both water regime treatments. Both irrigation and drought stress trials received 75 kg N ha -1 and 100 kg P ha -1 at sowing. At early stem elongation stage, 75 kg N ha -1 was also applied. To minimize other grain yield-reducing factors, fungicides were used to control diseases. Weed control was implemented in all stages of crop growth. The amount of water for each trial was calculated using the following formula [31]: Where, dn (g cm -2 ) is the height of irrigated water, FC is field capacity based on weight (%) of soil samples which was 33, 37 and 38% in 0-30 cm, 30-60 cm, and 60-90 cm soil depths respectively. ρb is soil bulk density (1.4 g cm -3 ), D shows soil depth (30 cm) and θm (%) denotes for soil moisture and calculated as follow:

Assay for proline concentration
Samples (0.5 g) of plant material were subjected to double extraction of free proline with 3% solution of 5-sulfosalicylic acid. The homogenate was centrifuged at 1000 g for 15 min. The proline content in the supernatant was determined by the method described in Bates et al. [32]. One ml of total supernatant was treated with 2 ml glacial acetic acid and 2 ml ninhydrin reagent (1.25 g ninhydrin dissolved in 30 ml of glacial acetic acid and 20 ml of 6 M phosphoric acid). The reaction was carried out for 1 h during incubation of samples in a boiling water bath. Next, samples were rapidly cooled on ice, mixed with 3 ml toluene and vigorously shaken. The absorbance of the pink-red toluene fraction was measured at 520 nm using toluene as a standard reference. Proline concentration was determined by comparing the absorbance data with the calibration plot for standard proline solutions (Sigma, United States). The content of free proline was expressed as mg g -1 of fresh weight.

Extraction Buffer
For preparing 100 ml of extraction buffer, 0.607 g of Tris-HCL was mixed with 50 g PVP in 90 ml water at a fixed pH 8. This buffer was used for the extraction of each antioxidant. Amount of 0.5 g leaf was completely powdered in liquid nitrogen and 2 ml of extraction buffer was added and homogenized completely in the Pounder mixture in a tube and centrifuged for 15 minutes at 13,000 rpm. At next stage, the supernatant was separated and enzyme activity was read spectrophotometrically.

Assay for catalase concentration
The content of catalase was determined by Dhindsa et al. [33] procedure. In this method, 50 μl of extract mixed with 50 mM potassium phosphate buffer (pH 7) and 15 mM hydrogen peroxide (H 2 O 2 ). The absorption was spectrophotometrically read at the 240 nm wavelength. One unit (U) of catalase decomposes 1 mM hydrogen peroxide (H 2 O 2 ) per minute. This relation was used for catalase quantification.

Assay for peroxidase concentration
Peroxidase activity was determined according to the method of Chance and Maehly [34]. The enzyme was assayed in a solution containing 50 mM phosphate buffer (pH 7.0), 5 mM H 2 O 2 and 13 mM guaiacol. The reaction was initiated by adding 33 μl enzyme extract at 25°C. One unit of enzyme was calculated based on formation of tetraguaiacol. Tetraguaiacol has a maximum absorption at 470 nm and its reaction can be readily followed spectrophotometrically.

Data Analysis
Analysis of variance (ANOVA) was performed using statements in SAS software and the traits means were statistically compared using the least significant differences (LSD) test. The correlations of grain yield and enzymes under drought conditions were also estimated.

Proline Variations as Non-enzymatic Reaction to Drought Stress
Effects of drought stress and cultivar and their interaction on proline accumulation were significant (Table 1). Drought increased proline accumulation (Fig. 1). Other report indicates that proline is one of the most critical osmolytes and drought causes a sharp increase in the accumulation of this amino acid in plants challenging with drought stress [35]. Variations in proline content in landraces and commercials are shown in Fig. 2 FW) proline accumulated in KC4863 but the lowest (1.69 U g -1 FW) was found in KC1948 (Suppl. Table 1). Proline accumulation is higher in drought-tolerant varieties compared to susceptible ones [36,37].
The highest accumulation of proline in normal conditions was observed in KC3885, KC4863, KC4144, KC4907, KC4511, KC4510, KC4880, KC4890, KC4528, KC4779, KC4529 and in the commercial cultivar Bezostaya. Some of these genotypes such as KC4510, KC4144, KC3885, KC4863, KC4907, and KC4528 and Bezostaya had also high proline under drought that shows their capability to react properly to water deficit conditions. Some of genotypes such as KC4838, KC1948, KC4632, KC3879, KC4637, KC4692, KC4700 and KC809 are categorized as susceptible because they reacted slowly to water deficit conditions and accumulated lower proline under drought. Regression model showed that proline and grain yield had positive association and increased proline content caused higher grain yield (Fig. 3). In many plant species, proline accumulation involves in osmotic mitigation and the protection of cellular apparatus in the lowyielding environments [38].

Variation in Enzymatic-Antioxidants in Response to Drought
Analysis of variance showed that a combination of effects including irrigation regimes, genotype and their interactions were responsible for changes in antioxidant accumulations. Fig. 4 shows that catalase accumulated more considerably in drought stress than in normal irrigation conditions.
Regression of grain yield on catalase content indicated that grain yield increased linearly as catalase activity increased (Fig. 6). The difference of catalase accumulation in normal and drought conditions was negligible in some genotypes such as KC1948, KC4692, KC4637, KC3879, KC4700, KC809 and Shiraz. Gong et al. [39], Zhu et al. [40] and Liu et al. [1] emphasized that an increase in antioxidant enzyme activity is corresponded to environmental stresses and type of plant genotype.  and Bezostaya also accumulated higher catalase under drought condition. This means these genotypes have high potential for fast reaction against water deficit conditions after heading via triggering antioxidant systems. KC4494, KC4641, KC4498, KC4870, KC4502, KC4880 had also high capability for catalase accumulation in response to drought occurrence. Gong et al. [39] reported that after 24 h of drought occurrence catalase activity was rapidly increased which shows the important role of this enzyme against adverse effects of drought stress.

leaf fresh weight under drought stress and normal irrigation conditions
Drought stress increased peroxidase accumulation significantly (Fig. 7). and catalase (X) as U g -1

leaf fresh weight in wheat under drought stress
Regression of grain yield on peroxidase activity showed that peroxidase had linear relation with grain yield (Fig. 9). High level of diversity in peroxidase accumulation in current wheat genotypes indicates that selection for drought tolerant is possible if it accompanies with higher grain yield.

Grain Yield Variations under Drought Stress
The results of ANOVA showed that irrigation, genotype and their interactions significantly affected grain yield (Table 1). Grain yield affected significantly by post-heading drought stress (Fig. 10). This means that from heading stage, wheat is highly sensitive to water deficit condition. The commercial cultivar of Shiraz (811.13 g m -2 ) and the landrace variety KC4703 (806.5 g m -2 ) had the highest grain yield in normal irrigation conditions (Suppl. Low -yielding landraces were KC4528, KC4890,  KC4880, KC3885, KC4498, KC4840, KC4502,  KC4494, KC4779, KC4529, KC4144, KC4907,  KC4641 and KC4510 with 75.9 to 189.4 g per square meter. Great variation was observed in grain yield of wheat genotypes in dealing with severe drought stress (Fig. 11). This shows the possibility of selecting high and low-yielding genotypes to be used in cross breeding programs for mapping genes responsible for drought tolerance. Results of Alaei et al. [41] and Nouri-ghanbalani et al. [42] studies indicated large differences in the sensitivity of wheat grain yield in dry-land and irrigated conditions. Nourighanblani et al. [42] reported that the performance of a genotype when it challenges with drought is influenced by a complex combination of morphological, physiological and phenological characteristics.

Fig. 11. Mean grain yield (vertical axis) as g m -2 for wheat genotypes (horizontal axis) under normal irrigation regime (Red color) and drought stress (blue color)
Therefore, screening and selection of wheat genotypes based on higher antioxidants does not affect grain yield under drought stress condition and using one of these antioxidants as indirect selection criteria for drought tolerance depends on the cost of antioxidant extraction.

CONCLUSION
The results of this study showed variation in the activities of antioxidant enzymes, proline and grain yield in wheat genotypes. Antioxidants were significantly increased under drought stress that shows they provide a defense line against adverse effects of free radicals released under drought stress conditions. Results also indicated that accumulation of higher antioxidants is related to higher grain yield production when wheat plants deal with water deficit condition specifically after heading stage. Proline which is a non-enzymatic antioxidant had significant association with higher grain yield and this shows it is an efficient character for selecting highyielding genotypes under drought conditions. KC4144, KC4641, KC4779, KC3885, KC4529, KC4863, KC4907, KC4528, KC4511 and KC4840 had higher enzyme activities. These genotypes had also higher average grain yield in both irrigation and drought stress conditions. The genotypes KC4830, KC1948, KC3879, KC4637, KC4700, KC809 and Shiraz varieties had lower enzymatic activity and categorized as susceptible.
Mapping populations are constructed mainly based on a cross between a high-yielding drought tolerant and a susceptible one in order to locate QTLs or genes responsible for drought-adaptive traits. Therefore, categorizing genotypes as drought-tolerant and susceptible would help a better programming of breeding for higher grain yield and stability under low-income environments.