Introduction

Wheat is a major crop worldwide and owing to its good adaptability, it is grown in almost all climate types. However, agricultural production is often threatened by unfavorable environmental factors of various kind (NASA). Drought in particular imposes significant peril for wheat production (Feller and Vaseva 2014). Dehydration causes water loss from tissues which leads to lowered cellular turgor, reduced cell elongation, and decreased growth and development at the whole plant level (Pandey et al. 2015). The pursuit of resistant and at the same time highly productive genotypes is a major goal for sustainable agriculture (Newton et al. 2010). Although some wheat varieties possess substantial productive potential, they are often exposed to drought periods which might lead to significant yield reduction (Bányai et al. 2020). Thus, along with higher yield capacity, drought tolerance is a desirable trait for modern wheat varieties. In this regard, wheat genotypes should be tested for their drought tolerance at early growth stages when they are most susceptible to damages caused by water deficiency.

Nevertheless, surviving in hostile environments requires complex abilities. Plant adaptations to water shortage could be recognized by different mechanisms among which drought escape and drought avoidance are the most prominent (Shavrukov et al. 2017). At the cellular level abiotic stress leads to enhanced generation of reactive oxygen species (ROS) and free radicals in plants which contribute to the development of oxidative stress. Aerobic organisms acquire energy through the use of molecular oxygen as final acceptor in the electron transport chain in the process of respiration. Although molecular oxygen per se is not highly reactive and is usually harmless it is capable of being partially reduced and can produce intermediates such as singlet oxygen (1O), superoxide (O2•) and hydroxyl (•OH) radicals and hydrogen peroxide (H2O2). Under optimal conditions ROS are produced in plants at low levels serving as signal transduction molecules that regulate stress response as well as growth and developmental processes but their accumulation drastically increases when plants are exposed to stress (Noctor et al. 2018; Mahmudi and van Breusegem 2018). Increased ROS accumulation negatively affects cell membrane stability by causing lipid peroxidation, damage nucleic acids, and proteins and thus inactivate and impair their proper functioning in the cell. In this regard, plants have developed a fine antioxidant defense system which is directly or indirectly associated with ROS and free radical scavenging and represents a powerful tool for neutralization of oxidative stress at the cellular level. Significant part of the defense is represented by enzymes, while a non-enzymatic component comprises molecules with high antioxidant capacities like phenols, thiols, and vitamins (Laxa et al. 2019; Ulricha and Jakob 2019). Thus, plants’ ability to tolerate stress is partly determined by the efficiency of the antioxidant defense system.

Improved growth and productivity of crops in general and wheat in particular is greatly dependent on the duration of stress and its time of appearance (Blum 2017). Therefore, more effective water use in the early vegetation period could contribute to yield improvement under drought conditions.

The aim of the present study was to evaluate drought responses of different wheat varieties at early seedling stage and to assess their tolerance in terms of survival under dehydration and subsequent rehydration. Special attention was drawn to plants’ ability to recover from stress which could be regarded as an important feature of acquired tolerance. Two newly established high yielding wheat varieties were compared with two standard varieties with contrasting agronomical drought resistance defined on the basis of their yield reduction after drought experienced in the field. Water retention capacity in the leaves, cell membrane stability, compatible solutes accumulation, and efficiency of the antioxidant protection system were assessed. By exposing differences in the reaction toward drought and the ability to recover we aimed to elucidate possible mechanisms for achieving drought tolerance and to select wheat varieties suitable for cultivation in drought-prone environments.

Materials and methods

Plant material, growth conditions, and treatments

Four Bulgarian common wheat (Triticum aestivum L.) varieties were chosen for assessment of drought tolerance at seedling stage. Jantar and Dobrudjanka represent drought tolerant and drought sensitive standards, respectively, based on the reduction of grain yield under drought experienced in the field. Iveta and Bojana are newly established high yielding varieties with unknown responsiveness toward field drought. Plants were grown as soil cultures in leached meadow cinnamon soil (pH 6.2, optimally fertilized with N, P, and K) under 190 μE.m−2 s−1 PAR at 21–25 °C, 16-h photoperiod and relative soil humidity of 70%. Soil humidity was monitored and controlled gravimetrically. Drought stress was imposed on 10-days-old plants with fully developed first leaf and an expanding second leaf by withholding irrigation for 7 days, followed by 3 days of rehydration during which watering was resumed. After dehydration soil water content dropped to as low as 20% of the maximal soil water retention capacity. Control plants were watered daily during all 20 days of the experiment. For all experiments, fresh leaf material was collected and used immediately or frozen in liquid nitrogen and stored at—70 °C until measurement.

Relative water content and cell membrane damage

Relative water content (RWC) was measured in the second fully developed leaf and was calculated according to the equation: \(\mathrm{RWC} =\left(\frac{\mathrm{FW}-\mathrm{DW}}{\mathrm{TW}-\mathrm{DW}}\right)\times 100\),

where FW represents fresh weight, TW is weight at full turgor, measured after incubating the leaves in water for 24 h, and DW is weight after drying the leaves at 80 °C until reaching constant weight.

Cell membrane damage was estimated based on conductivity measurements of electrolyte leakage from second leaf samples. Leaves were cut into 2 cm-long pieces which were immersed in distilled water and incubated for 24 h at 10 °C in the dark (15 leaf pieces in 15 ml of distilled water). Conductivity of the solutions was measured twice with Hanna Instruments (USA) conductometer: first measurement was performed after 24 h of incubation (κ) followed by a second measurement after boiling the samples for 10 min (κmax) which represented the total electrolyte content. Membrane damage was expressed as percentage of leaked compared to the total electrolytes given by the relation \(\frac{k}{k\mathrm{max}}\times 100\).

Biochemical analysis

Selected parameters such as content of free proline, malondialdehyde, total phenols, thiol-containing compounds, hydrogen peroxide, total soluble protein and activities of catalase, guaiacol peroxidase, and superoxide dismutase were measured. Fresh leaf material (approximately 300 mg) was homogenized with 0.1% (w/v) trichloroacetic acid for determination of free proline, soluble phenols, hydrogen peroxide (H2O2), malondialdehyde (MDA), and free thiol-containing compounds.

Free proline was extracted, derivatized with acid ninhydrin, and absorbance was read at 520 nm according to the method of Bates et al. (1973). Hydrogen peroxide content was estimated spectrophotometrically at 350 nm according to Alexieva et al. (2001). The amount of hydrogen peroxide was calculated using a standard curve prepared with known concentrations of H2O2.

Malondialdehyde content was determined as thiobarbituric acid-reagent product according to Kramer et al. (1991) using the extinction coefficient of 155 mM−1 cm−1. Total phenolics content was determined in leaf extract in 0.1% (w/v) trichloroacetic acid with Folin–Ciocalteu reagent supplemented with sodium carbonate and absorbance was read at 725 nm according to the method of Swain and Goldstein (1964). Gallic acid was used as reference standard. Spectrophotometric measurements were taken on Multiskan Spectrum (Thermo Electron Corporation, Finland). Free thiol-containing compounds (SH groups) were determined with Ellman’s reagent and absorbance was read at 412 nm (Edreva and Hadjiiska, 1984).

Measurement of antioxidant enzyme activities. Fresh plant material (500 mg) was homogenized at 4 °C in 100 mM potassium phosphate buffer (pH 7.0) containing 1 mM EDTA and 1% polyvinylpyrrolidone (w/v). The homogenates were centrifuged at 12 000 × g for 15 min. Soluble protein in the supernatant was determined by the dye-binding technique (Bradford 1976) using bovine serum albumin as standard. Catalase (CAT, E.C. 1.11.1.6) activity was measured following hydrogen peroxide decomposition by monitoring absorbance decrease at 240 nm (ε = 36.8 mM−1 cm−1) for 1 min (Aebi 1984). Superoxide dismutase (SOD, E.C. 1.15.1.1) activity was measured according to Beauchamp and Fridovich (1971). Ascorbate peroxidase (APX, E.C. 1.11.1.11) activity was determined according to Guo et al. (2007). Guaiacol peroxidase (GPX, EC 1.11.1.7) activity was measured in a reaction mixture containing 50 mM phosphate buffer (pH 7), 9 mM guaiacol, and 19 mM H2O2 (Lin and Kao, 1999). Activity was calculated using the extinction coefficient (26.6 mM−1 cm−1) at 470 nm. One unit of peroxidase was defined as the amount of enzyme that caused the formation of 1 mM of tetraguaiacol per 1 min. The kinetic measurements of CAT, APX, GPX, and SOD activities were taken on spectrophotometer Shumadzu UV-1601 UV–Visible (Shimadzu, Japan).

Statistical analysis

Two independent experiments were conducted with at least three technical replications for each parameter. Data were expressed as means ± SD. Significance of differences was analyzed by XLSTAT Version 2014.5.03 and Duncan’s multiple range test was applied at significance level P < 0.05.

Results

When the four wheat varieties were grown on sufficient water supply, they demonstrated full leaf hydration. Drought imposed by withholding water for 7 days caused significant reduction of leaf RWC in all studied genotypes with highest values observed in Iveta and Jantar and lower in Bojana and Dobrudjanka (Fig. 1a). Upon re-watering, hydration was restored but only in Dobrudjanka control values of RWC were reached and Bojana reached only half of its control turgidity. Leaf-free proline content in controls was commensurable in all four wheat varieties, but under drought it increased greatly. Highest amounts of this imino acid were accumulated in Dobrudjanka, followed by Bojana and Iveta, and lowest in Jantar (Fig. 1b). Upon rehydration proline levels decreased but Dobrudjanka sustained highest values among the varieties. Another parameter that increased significantly under drought in all varieties was malondialdehyde content. Greatest amounts of MDA were found in Dobrudjanka (more than 20-fold increase over controls), followed by Bojana, Iveta, and smallest in Jantar (Fig. 1c). Upon re-watering MDA levels decreased in the four varieties most notably in Dobrudjanka but for all genotypes it remained higher than respective controls.

Fig. 1
figure 1

Physiological parameters measured in the leaves of four wheat varieties under sufficient water supply and subjected to drought and subsequent re-watering (rehydration): a relative water content (RWC); b free proline; c malondialdehyde (MDA) and d hydrogen peroxide (H2O2). Values are means ± SE (n = 6). Different letters represent significant differences at P < 0.05

Full size image

Hydrogen peroxide content under sufficient water supply was highest in Jantar and lowest in Bojana, while the other two varieties, Dobrudjanka and Iveta, had similar intermediate levels of this solute (Fig. 1d). Drought caused greatest increase in H2O2 in the leaves of Dobrudjanka (almost nine times higher than controls) and lowest in Jantar (twofold exceeding control values) while Bojana and Iveta had sixfold and fivefold increase in H2O2 compared to respective controls. Upon rehydration, hydrogen peroxide levels decreased most prominently in Dobrudjanka, reaching control values in Iveta and Jantar and remaining more than two times higher than controls in Bojana.

Highest electrolyte leakage was observed under drought in varieties Bojana and Jantar representing 11- and eightfold increase over respective controls (Table 1). In Dobrudjanka twofold increased leakage was observed and in Iveta drought led to four times higher leakage than controls. Upon rehydration, membrane leakage was recovered only in Dobrudjanka, while in the other three varieties electrolyte leakage was greater than controls, most prominently observed in Bojana, followed by Jantar and Iveta.

Table 1 Percentage of estimated membrane damage in the leaves of four wheat varieties under sufficient water supply (control) and subjected to drought and subsequent re-watering (rehydration). Values are means ± SE (n = 6). Different letters represent significant differences at P < 0.05
Full size table

Drought caused most dramatic increase in SOD activity in Dobrudjanka (more than 12-fold) and only twofold in Jantar (Fig. 2a). In Bojana SOD increased slightly upon dehydration and did not change in Iveta. After re-watering SOD activity returned to control levels in all varieties except Dobrudjanka, where it remained almost twice higher than respective control. CAT activity increased drastically in Jantar and Bojana (more than eightfold exceeding well-watered controls), but decreased slightly under drought in Dobrudjanka and Iveta (Fig. 2b). Upon re-watering CAT activity returned to control levels in the last two variants, but remained higher than controls in Jantar and Bojana.

Fig. 2
figure 2

Enzyme activity in the leaves of four wheat varieties under normal water supply (control) and subjected to drought and subsequent re-watering (rehydration): a superoxide dismutase (SOD); b catalase (CAT); c ascorbate peroxidase (APX) and d guaiacol peroxidase (GPX). Values are means ± SE (n = 6). Different letters represent significant differences at P < 0.05

Full size image

In well-watered plants APX activity was lower in Iveta and Dobrudjanka and higher in Jantar and Bojana (Fig. 2c). Drought caused decrease in APX activity in Dobrudjanka, and slight increase in Bojana, while in the other varieties it led to only insignificant changes. Upon re-watering APX showed higher activity compared to controls in Dobrudjanka and Jantar and lower activity in Iveta, while in Bojana control and rehydrated plants had similar APX activity.

GPX activity in the leaves of well-watered plants was highest in Jantar, lower in Bojana and Dobrudjanka and lowest in Iveta (Fig. 2d). Greatest increase in GPX activity under drought conditions was found in Jantar, while the other three varieties (Dobrudjanka, Iveta, and Bojana) did not exhibit significant changes regarding the activity of this enzyme.

Control plants from all four wheat varieties had similar amounts of phenolic compounds in the leaves and drought caused significant increase in these substances in all variants (Fig. 3a). Highest amounts of phenolics were observed in Dobrudjanka (more than tenfold increase), followed by Iveta and Bojana (almost fivefold increase), and lowest, threefold increase was found in Jantar. Upon rehydration, phenolics content decreased although remaining between two and four times higher than controls with greatest values assessed in Iveta and lower in Bojana and Dobrudjanka, followed by Jantar.

Fig. 3
figure 3

Amounts of (a) phenols and (b) sulfhydryl (SH) groups in the leaves of four wheat varieties under normal water supply (control) and subjected to drought and subsequent re-watering (rehydration). Values are means ± SE (n = 6). Different letters represent significant differences at P < 0.05

Full size image

Lowest content of SH groups was measured in Dobrudjanka and Iveta while the other varieties had similar lower amounts of thiols under well-watered conditions (Fig. 3b). Drought caused increase in SH groups in all varieties which was most pronounced in Dobrudjanka (six times above controls) and Iveta (five times higher than control values) and to a lower extent in Jantar and Bojana. Upon re-watering, content of free thiols decreased in all varieties except Bojana, but remained higher than corresponding control levels for all variants.

Discussion

Water retention capacity under drought conditions is a desirable trait and often correlates with drought tolerance provided that it is accompanied by maintaining higher yield (Marcek et al. 2019). Relative water content (RWC) in the leaves is an important physiological feature considered to provide satisfactory concept of the relationship between plant water status and its metabolic activity (Soltys-Kalina et al. 2016). It is a sensitive parameter that reflects changes in water relations in response to water shortages in the environment and relates physiological processes in the plant to soil water status, therefore, representing a reliable measure of water retention capacity (Waraich and Ahmad 2010). In our experiments stress induced by water deprivation could be viewed as moderate (non-lethal) in strength and duration and provided better physiological response in contrast to severe stress which would lead to irreversible damages in the plants. The presented results revealed that wheat varieties Jantar and Iveta preserved higher RWC values under drought compared to Dobrudjanka and Bojana. However, upon re-watering only Dobrudjanka fully recovered its leaf water content to control levels although it was recognized as the most sensitive to drought periods in the field.

Plants minimize the harmful consequences of water deficiency through the process of osmotic adjustment by enhanced accumulation of compatible solutes, also called osmolytes, such as free proline, glycine betaine, soluble sugars, and polyols. Such compounds contribute to the development of drought tolerance in plants (Ahmed et al. 2019). Proline seems to be the most typical osmolyte although its action could not merely be restricted to osmoprotection but brings about stabilization of membranes, proteins, and other subcellular structures under osmotic stress (Kaur and Asthir 2015; Wang et al. 2019). Proline also acts as metal chelator, antioxidant protector, and signal molecule. Stress-induced proline accumulation probably contributes to the maintenance of cell turgor and the stabilization of cellular membranes, as well as balancing of ROS and preventing oxidative damage. Positive correlation was established between proline accumulation and the strength and duration of water stress (Blum 2017). It was clearly demonstrated that proline content in several wheat varieties was not a direct marker for drought tolerance although its osmoregulatory role was undisputed (Marcek et al. 2019). In our experiments drought-susceptible genotype (Dobrudjanka) had highest increase in proline content, while on the contrary, drought-resistant Jantar had lowest proline content under drought stress. It could be speculated that accumulated free proline was useful in recovery processes as C and N sources for resuming normal metabolic processes upon stress relief (de Sousa et al. 2020). Obtained here results also support the probability of the above hypothesis.

Under optimal conditions cellular metabolic activity sustains an equilibrium between the accumulation and elimination of ROS. However, during their life cycle plants are often exposed to stress influences leading to excessive production of ROS such as singlet oxygen (1O2), superoxide radical (O2•), hydrogen peroxide (H2O2), and hydroxyl radicals (OH). Their over-production is closely related to the formation of MDA and represents a convincing biochemical marker for the development of oxidative stress. Recently, a similar dependency was evidenced in wheat plants subjected to soil drought (Todorova et al. 2021). Accumulation of MDA was observed in response to various stress factors and was recognized as indicator of lipid peroxidation causing loss of membrane integrity and function (Hasanuzzaman et al. 2020). Impaired membrane permeability consequently leads to increased electrolyte leakage from damaged tissues (Kocheva et al. 2014; Öztürk et al. 2016). We found that in drought-susceptible wheat variety Dobrudjanka higher levels of MDA were consistent with greater membrane damage which could not be alleviated by highest proline levels accumulated in this genotype under dehydration.

It is well known that hydrogen peroxide could inhibit the activity of many enzymes including metaloproteins like SOD. On the other hand, ROS could have beneficial role in plants (Raja et al. 2017). The most dramatic drought-induced increase in stress markers (MDA, proline, and H2O2) among studied wheat varieties was documented in Dobrudjanka. This increase was in consonance with the well-established fact that drought provokes the generation of ROS which diffuse through biomembranes and cause serious cellular damage and even lead to cell destruction. Such increment of stress markers has been well documented in wheat (Abid et al. 2018; Amoah et al. 2019). In some studies, enhanced antioxidant accumulation correlated with higher stress tolerance and contributed to greater antioxidant capacity (Mattos et al. 2015). In our experiments, the preferred mechanism of oxidative stress response in the drought tolerant wheat variety Jantar was through activation of the enzymatic component of the antioxidant defense system, while in drought sensitive Dobrudjanka accumulation of non-enzymatic compounds like phenolics, thiols, and proline was employed for assuring stress protection. Thus, the two contrasting genotypes implemented different strategies for overcoming and surviving the stress period. Upon re-watering, accumulated proline and antioxidant compounds most probably served in the alleviation of stress and the more complete recovery of leaf water balance in Dobrudjanka.

Decreased photosynthetic activity under drought leads to lowered metabolic efficiency and serious reduction in growth processes and eventually contributes to the reduction of crop yield (Guo et al. 2007; Marek et al. 2019). Therefore, it could be hypothesized that varieties with better water use efficiency under dehydration would exhibit superior drought tolerance which could contribute to higher grain yield. Recovery from drought stress is a dynamic process of restoring cellular functions, reconstitution of damaged structures and return to normal metabolic activities which is of chief importance for plant stress adaptability (Kirova et al. 2021). It was demonstrated earlier that wheat plants responded to water deficiency through reduction in transpiration rate and vegetative growth, and lowered stomata number per leaf area, which was regarded as a representation of drought escape mechanism (Shavrukov et al. 2017).

Plant acclimations to environmental changes require a shift of homeostasis which can be achieved through delicate balance of multiple metabolic pathways localized in various cellular compartments (Scheibe 2019). SOD is a powerful antioxidant enzyme present in all aerobic organisms that secures first-line defense against increasing ROS amounts. Stimulated SOD activity increased stress tolerance in wheat plants (Lakhdar et al. 2010; Sen et al. 2017). Significantly increased antioxidant enzymes activity in leaves was detected under various abiotic stresses like drought, salinity, and heavy metal toxicity in many plant species (Al Hassan et al. 2017; Kapoor et al. 2019; Nasirzadeh et al. 2021). CAT possess the potential for direct dismutation of H2O2 to water and oxygen. Along with SOD, peroxidases and catalases form plant defense system, assist the removal and detoxification of superoxide radicals (Garg and Manchanda 2009) and are also important for H2O2 generated in peroxisomes as side product of the oxidation of fatty acids and photorespiration. Chakraborty and Pradhan (2012) demonstrated that wheat varieties exposed to drought had initially increased SOD and CAT activity which subsequently decreased upon relief of stress. Similarly, in our experiments, drought induced great increase in CAT, APX, and GPX activities in the tolerant wheat variety Jantar and in Bojana and highest SOD levels in the sensitive variety Dobrudjanka and these activities tended to returning to control values after stress was dismissed.

Antioxidants are molecules which significantly delay or prevent the oxidation of a certain substrate even at low concentrations (Halliwell 2006). Phenols are excellent antioxidants due to the capacity of their acid hydroxyl group to donate electrons. Many natural and synthetic antioxidants like tocopherol (vitamin E) are in fact phenolic compounds with ROS scavenging capacity (Raal et al. 2015). They can directly remove toxic ROS which can easily oxidize phenols to phenoxyl radicals (Herrmann et al. 2015). A tight relationship exists between drought and the induction of oxidative stress through enhanced accumulation of ROS which alters antioxidative pathways in plants aimed toward stress management (Talaat et al. 2015). Liu et al. (2018) demonstrated that phenolic compounds contributed to abiotic stress adaptation due to their antioxidant activity and under drought conditions stress-tolerant wheat varieties had higher phenol levels in the leaves than sensitive ones. Our results were not in consonance with these findings but rather supported the idea that these compounds could be useful for recovery processes following stress relief.

Differences in the metabolic responses to drought between tolerant and susceptible wheat genotypes were recently described and sensitive wheat varieties were represented as experiencing higher degree of oxidative stress (Simova-Stoilova et al. 2009; Tsenov et al. 2011; Marek et al. 2019;). In our experiments the most tolerant wheat variety Jantar was able to successfully withstand dehydration and by maintaining highest leaf hydration, lowest membrane damage and highest enzymatic activity. Based on their overall performance under drought and re-watering, the response of newly established wheat variety Bojana to drought was more similar to Jantar while the rection of Iveta was more similar to Dobrudjanka. Considering all measured physiological and metabolic responses together, wheat varieties Iveta and Bojana could be regarded as moderately drought tolerant since the intensity of their reactions toward imposed dehydration was intermediate between those of tolerant Jantar and sensitive Dobrudjanka.

Future experiments would be required to determine whether established performance in laboratory conditions would also be beneficial for preserving higher yield under drought periods in the field. The presented results suggest that although generally/commonly used wheat genotypes demonstrate relatively narrow genetic variation, they might have developed different strategies or coping mechanisms for withstanding stress conditions. Breeding programs would profit from the establishment of quick and convenient laboratory methods for prediction of drought tolerance which could be useful in the selection of genotypes suitable for cultivation in drought-prone regions.

Nevertheless, we are aware that our results should be regarded with caution, in the sense that drought tolerance is a highly complex feature and is not necessary equivalent in different vegetative stages. Comparison of drought resistance should be accomplished in corresponding developmental stages. Moreover, it should be emphasized that agronomically assessed drought resistance would not totally overlap with experimentally evaluated physiological tolerance.