Physiological alterations and enzymatic evaluation of soybean cultivars under water deficit

Drought is one of the main abiotic factors limiting agricultural productivity, capable of having a major impact on the yield of most crops. The knowledge of the physiological and biochemical mechanisms that differentiate resistance and susceptibility to water deficit among soybean strains can be used in the generation of more tolerant cultivars. In this sense, the objective of this study was to characterize physiologically, two soybean cultivars with different patterns of tolerance to drought in the field, by determining photosynthetic rates, lipid peroxidation levels and antioxidant enzyme activity under three levels of water potential. Upon reaching the V4 development stage, the irrigation of the plants was suspended and three data collection were performed: full irrigation (control); moderate water deficit (Ψ = -1.5 ± 0.2 MPa) and severe deficit (Ψ = -3.0 MPa ± 0.2 MPa). Variations in perspiration rate, stomatal conductance, as well as decrease in photosynthetic rate were significant between the two cultivars, where the water potentials in cultivar BR 16 anticipated on average two days achieving the same water potentials in Embrapa 48 cultivar, thus presenting better efficiency in water use. In addition, the increased activity of enzymes and lipid peroxidation were more significant in the cultivar BR 16, demonstrating that this cultivar is less tolerant to drought than Embrapa 48 cultivar, corroborating to agronomic data previously found in the field.


INTRODUCTION
Soybean plants are affected by various abiotic stresses such as high salinity, low temperatures, mineral element toxicity, drought and others (Ferreira et al., 2020;Escalera et al., 2021;Gontijo et al., 2021;Pimentel et al., 2021). Drought or continuous water deficiency is one of the most important factors affecting growth, development, survival and productivity of the crop (Manavalan, Guttikonda, Tran, & Nguyen, 2009;Fried, Narayanan, & Fallen, 2019). According to Food and Agriculture Organization of the United Nations (FAO) (2015) the international research community should be aware to solutions that reduce the problems caused by abiotic factors in agriculture, including drought. They also suggest that one of the possible solutions is the development and use of new plant cultivars.
Therefore the development of new cultivars tolerant to abiotic stresses, particularly to water stress, has a greater importance in soybean breeding programs. There are several studies related to this feature in Soybean, useful for these programs and for this purpose (Mwenye, Rensburg, Biljon, & Merwe, 2018;Fried et al., 2019;Iqbal et al., 2019). Recently, the use of biotechnological tools has favored more detailed information involving the identification, interaction and quantification of genes involved in Soybean water deficit tolerance (Langridge & Reynolds, 2015;Cilliers, Heerden, Kunert, & Vorster 2018;Khan et al., 2018;Ye et al., 2020).
Plant physiological responses to drought are of physiological, biochemical, morphological and molecular in nature, and include stomatal closure, decreased photosynthetic activity, alteration of the cell wall elasticity, membrane fluidity and generation of toxic metabolites causing the death of the plant (Ramírez, Querejeta, & Bellot, 2009). According to Xu et al. (2018) there are a large number of differentially expressed genes and the various pathways indicate that soy uses complex mechanisms to handle drought. However, they claim that some identified genes and pathways can be used in soybean breeding tolerant to water stress. It is therefore important to elucidate the processes occurring in plants under such conditions. Many of the deleterious processes supported by plants under water stress conditions are mediated by reactive oxygen species (ROS). The production of ROS can trigger the lipid peroxidation process in cell membranes, forming lipid hydroperoxides that lead to decreased fluidity, modifications of ionic permeability and other membrane-associated functions, thus being one of the most significant events of oxidative stress (Anjum et al., 2015;Anjum et al., 2017).
To reduce the damage caused by ROS, efficient enzymatic and non-enzymatic antioxidant defense systems act in a coordinated manner under stressful conditions in order to maintain intracellular homeostasis. The enzymes ascorbate peroxidase (APX), Glutathione peroxidase (GPX), catalase (CAT), Glutathione reductase (GR) and phenol peroxidase (POX), subsequently detoxify H2O2 by releasing H2O by different oxidation processes (You & Chan, 2015).
Understanding how plants respond to water deficit and understanding tolerance mechanisms is critical to predicting impacts on crop production and is currently one of the largest research topics for the development of more tolerant and productive cultivars. Thus, the physiological characterization may be an important and fast procedure to select different cultivars that in the same experimental condition may show different levels of tolerance to stress (Atkin & Macherel, 2009).
Within this context, the present study aimed at characterizing physiologically two conventional soybean cultivars, contrasting for drought tolerance, in order to better understand the mechanisms of response to this tolerance.

Plant material, cultivation conditions and experimental design
The experiment was conducted in the greenhouse of the Molecular Physiology Laboratory of the Federal University of Viçosa, during the year of 2016, under partially controlled conditions with average temperatures of 25°C and 60% humidity, with a variation of up to 15% more or less. Embrapa 48 and BR 16 soybean cultivars were selected because they present contrasting responses in the field when submitted to periods of water deficit, and the Embrapa 48 being considered the most tolerant to drought (Texeira et al., 2008).
The experimental design adopted was completely randomized with a 2 X 3 factorial scheme (2 cultivars x 3 water treatments), with five replications. The experimental unit consisted of one pot containing 6.5 kg of substrate / pot with 50% vermiculite and 50% washed coarse sand composition, and two plants of each cultivar, totaling 30 pots for the entire experiment.
After the emergence of the fully developed fourth trifoliate, irrigation was suspended and the cultivars were subjected to two water deficit levels, with daily measurements between 05:00am and 06:00am o'clock. The first level was considered moderate water stress, with water potential of Ψw = -1.5 Mpa. The second, as severe water stress, Ψw = -3.0 Mpa. The water potential of each plant was measured in a Scholander pressure chamber. For these evaluations, a third node leaf was used. For this, two leafs of each plant was used to water system adopted. Fourth node leaf collections were performed up to 12 days after the application of the water deficit condition for further physiological analysis.

Evaluation of physiological characteristics
Gas exchange determinations were performed with a portable photosynthesis meter, IRGA; portable model LI-6400xt, LI-COR Biosciences Inc., Lincon, Nebraska, USA. Measurements were always made in the median area of the fully expanded leaves, totally exposed to solar radiation in the period from 08: 30am to 10: 00 o'clock. The following characteristics were measured: photosynthetic rate (A) (µmol m -2 s -1 ); leaf transpiration rate (E) (mmol m -2 s -1 ); stomatal conductance on leaves (Gs) (mol m -2 s -1 ) and CO2 concentration in the intercellular spaces (A) (mmol m -2 s -1 ).
The level of membrane lipid peroxidation in leaf tissues was measured in terms of the malondialdehyde content formed (MDA, a lipoperoxidation product), determined by the thiobarbituric acid reaction (TBA) according to the method described by Gomes-Junior et al. (2006).

Statistical analysis
The data obtained was submitted to presuppositions based on normality and homogeneity of residual variances. After, the characters were submitted to variance analysis of 5% probability by F test. Afterwards the analysis of variance was carried out in order to identify significant interaction between treatments and cultivars, these significant were dismembered and significant differences between treatments were detected using the Tukey mean test (p <0.05).
The relationship between treatments was evaluated by multivariate analysis, and the main components (ACP) were obtained from MDA data and antioxidant enzymes. The grouping of the treatments was done by Tocher's methodology. The variables were analyzed using R software 3.5.2 version (R Core Team, 2018), with the aid of Cluster, RCMDR and Ade4 packages (Dray et al., 2018).

RESULTS AND DISCUSSION
According to the classification of experimental precision (coefficient of variation "CV%") proposed by Gomes (1990), the values found for the experiments of physiological parameters and activity of antioxidant enzymes, obtained a medium experimental precision (10.7% and 20.02%, respectively), which is favorable, as it strongly attests to the inferences raised. The days after the suspension of irrigation experiment showed a low experimental precision (CV%= 33.96). The high CV% can be explained by the fact that it is a variable determined by a quantitative characteristic, determined by many genes, as there is a high environmental influence. One way to reduce this effect is to increase the number of repetitions in the experiment and also opt for more rigid experimental designs, in order to mitigate the environmental influence.
After water restriction, it was observed that the cultivars reached moderate and severe water potentials at different times, and the tolerant cultivar had longer time to reach these water potentials. The cultivar Embrapa 48 required two days longer than the cultivar BR 16, indicating that this cultivar is more tolerant to drought (Figure 1). Several effects due to water deficit have been reported in plants, such as increased stomatal resistance, reducing leaf transpiration leading to lower CO2 availability for photosynthesis (Zhou, Lam, & Zhang, 2007). In the present study, photosynthetic and transpiratory rates, as well as stomatal conductance decreased significantly in relation to irrigated treatment among cultivars (Figure 2). These effects ultimately reduce the productivity of various crops such as Alfalfa (Medicago sativa L.) (Liu, Wu, Ge, Han, & Jia, 2018); Sunflower (Helianthus annuus L.) (Hussain, et al., 2018) and Soybean (Glycine max) (Sabagh et al., 2018;Gavili, Moosavi, & Haghighi, 2019). Although the dehydration curve shows that the cultivar Embrapa 48 delayed dehydration at different water deficit levels, in the absence of stress, there were no significant differences between cultivars ( Figure  2). Showing that under irrigated conditions the plants have similar physiological profiles. In addition, the tolerant cultivar had lower leaf water potential the day before the experiment, In addition, the tolerant cultivar had a lower leaf water potential the day before the experiment, which may be explained by the higher stomatal conductance and transpiration rate, suggesting that the tolerant cultivar had higher hydraulic conductivity. According to Gray et al. (2016) the combined effects on stomatal conductance, with factors such as temperature, in soybean may not change yield under different CO2 fertilization conditions, and higher doses of CO2 are not advantageous. It is proposed that during drought years, reduced stomatal conductance will ease the increased vapor pressure deficit in transpiration helping to preserve crop yield (Gorthi, Volenec, & Welp, 2019).
Except for the intercellular concentration of CO2, the differences found were due to the level of water restriction imposed, as well as by the cultivar. Transpiration rate, stomatal conductance and photosynthesis rate were strongly affected by water restriction levels and the increase in water potential induced a decrease in the real value of the measured variables. It is noteworthy that the largest variations were identified in cultivar BR 16, while Embrapa 48 showed greater resistance to the decrease of the parameters mentioned.
The non-significance of the values for the ratio between internal and external CO2 concentrations (Ci/Ca) indicates that the decrease in photosynthesis rate was due to increased stomatal resistance and also to the effect of water stress on photosynthesis. The gradual decrease of photosynthesis was smaller in the tolerant cultivar under moderate deficit, which may indicate inhibition of the most intense biochemical or photochemical phase in the sensitive genotype. According to Iqbal et al. (2019) the reduction of photosynthesis results in the reduction of CO2 diffusion in leaves due to lower internal and stomatal conductance.
The photosynthetic rate decreased from 55% and 95% in moderate and severe water potentials for cultivar BR 16, while in Embrapa 48 for the same water potentials, the rates were 15% and 78%. The transpiration rate and stomatal conductance parameters also have similar patterns, since these parameters directly affect the photosynthetic rate. In this sense, it is assumed that the cultivar Embrapa 48 can tolerate more the imposition of water stress in relation to cultivar BR 16, maintaining higher photosynthesis levels under the same water potentials.
The gradual decrease of photosynthesis was smaller in the tolerant cultivar under the two water deficit levels, with differences in stomatal conductance, which may indicate inhibition of the most intense biochemical or photochemical phase in the sensitive genotype. The drastic reduction in stomatal conductance identified in cultivar BR 16 may possibly have led to decreased consumption of electrons released from water, causing excess energy that reacts with oxygen, causing oxidative stress and initiating the process of lipid peroxidation in cell membranes and detoxification by antioxidant enzymes (Apel & Hirt, 2004;Guo, Tian, Liu, Wang, & Sui, 2018).
The higher tolerance to drought of cultivar Embrapa 48 can also be verified in lipid peroxidation (MDA) rates. Although both cultivars have increased oxidative damage with the imposition of water stress, it was always higher in tolerant leaves in all treatments. Overall, the increase in MDA levels was proportional to the level of stress imposition ( Figure 3A). For the water potential level ψma = -1.5Mpa, the increase was greater in BR 16 (106%), while in Embrapa 48 the increase was only 56%. Several studies illustrate the increase in lipid peroxidation under water stress: in Sweet corn (Zea mays) (Terzi, Güler, Güven, & Kadioglu, 2018); Rice (Oryza sativa) (Qureshi et al., 2018;Gorthi et al., 2019) and Sorghum (Sorghum bicolor) (Guo et al., 2018).
The increase in activity of antioxidant enzymes was higher for the water potential Ψ = -1.5 MPa, compared to the irrigated one. In cultivar BR 16 the increase in enzyme activity was greater than the variation identified in Embrapa 48. In the first cultivar, the increase rates were 71% for GR enzyme, 73% for SOD, 57% for CAT, 90% for APX and 123% for POX. In the cultivar Embrapa 48 the increase was 19% for 24% GR, 20% for SOD, 33% for CAT, 14% for APX and 64% for POX, in relation to the rates found in irrigated treatments (Figure 3).
The consequences of the ideal lack of water are oxidative stress and reduced photosynthetic characteristics (Guo et al., 2018). Oxidative stress stimulates biosynthesis of antioxidant components and increases the activity of antioxidant enzymes. However, some species or cultivars may have greater resistance to this stress, contributing to a less pronounced increase than in other plants. Its concentrations and activities are related to many physiological processes involved in cellular signaling mechanisms in plant defense or oxidative stress (Soares & Machado, 2007).
Thus, the higher antioxidant activity in cultivar BR 16 can be explained by the fact that it has a lower ability to drain excess of reducing power from oxidative stress such as photorespiration, membrane peroxidation, cyclic electron flow in the thylakoid, among others. Therefore, a significantly larger increase is needed to counteract the effects of drought-induced oxidative stress (Scheibe, Backhausen, Emmerlich, & Holtgrefe, 2005;Heber, 2008).
It was observed, with the study of the similarity pattern between cultivars, that the first two main components explained approximately 97.34% of all observed variation, which allowed the dispersion of treatments in the Cartesian plane quite reliably (Figure 4). It was found that the most similar treatments were grouped according to the imposition of water stress.
The six treatments were associated in three distinct groups. Group I formed by the irrigated treatments. Group II by Embrapa 48 in the two levels of stress imposition and Group III formed by the two treatments in BR 16. In conclusion, regardless of cultivar, in the absence of water stress, both cultivars respond in a similar way. On the other hand, for the other two stress levels, (ψma) = -1.5Mpa and (ψma) = -3.0Mpa, the grouping was dependent on the cultivar, so stress imposition varied within each cultivar and was not similar for the same level of water stress in the different cultivars.

CONCLUSIONS
Both cultivars responded differently to the imposition of water stress on the physiological parameters analyzed. The cultivar Embrapa 48 proved to be more tolerant to this stress than the cultivar BR 16. The data allowed us to suggest that such physiological parameters can be used in the evaluation and distinction of tolerant and water deficit sensitive soybean genotypes. Escalera, R. A. V., Carvalho, I. R., Pimentel, J. R., Troyjack, C., Szareski, V. J., Jaques, L. B. A., … Pedó, T.