Responses of maize hybrids to water stress conditions at different developmental stages: accumulation of reactive oxygen species, activity of enzymatic antioxidants and degradation in kernel quality traits

Experimental material and site

This experimental study was conducted at the research area of Maize and Millets Research Institute, Yusafwala-Sahiwal (MMRI), Pakistan (longitude 73°12′39.3″E, latitude 30°40′57.1″E, altitude 173 m) under field conditions for two consecutive spring seasons (2019–2020 and 2020–2021). Seeds of three hybrids (YH-5482, YH-5427, and YH-5395) were obtained from the maize hybrid development group, Maize and Millets Research Institute, Yusafwala, while JPL-1908 was obtained from Jullundur Pvt. Ltd., Arifwala, and NK-8441 from Syngenta Seeds (Pakistan) respectively, during spring 2019–2020. About 1 kg seed per hybrid was stored at very low temperature and controlled humidity conditions in a cold store, installed at MMRI, Yusafwala to use in the next spring season (2020–2021). The field selected for the study had well-aerated, loamy sand textured soil having pH 7.8 and 16.51 g kg⁻¹ soil organic matter.

Water stress treatments

Maize hybrids were sown during mid-February under four conditions. (a) control (12 irrigations); (b) water stress at the pre-flowering stage (10 irrigations, two consecutive irrigations were held from 40 days after sowing); (c) water stress at the anthesis stage (10 irrigations, two consecutive irrigations were held from 60 days after sowing); (d) water-stress at anthesis stage (10 irrigations, two consecutive irrigations were held from 80 days after sowing). Hybrids were planted under randomized complete block design under split plot arrangement in triplicates. All other standard agronomic practices i.e., seed treatment, application of pre-emergence pesticide, insecticide for the control of shoot fly and maize borer. Recommended fertilizer application was given to all stress treatments at 92 kg nitrogen, 58 kg phosphorus, and 37 kg potash, respectively.

Collection of samples for biochemical analysis

Three weeks after the stress was applied, three fully expanded, healthy cob leaves were taken from each replication. Following a distilled water wash, the leaves were frozen in liquid nitrogen (N₂) and kept at −80 °C for the analysis of several biological components.

Measurement of yield-related morpho-agronomical traits

Phenological traits i.e., days to 50% anthesis and days to 50% silking were measured as the days taken by the plant to complete their 50% anthesis and silking from date of sowing. Similarly, several morphological traits including plant and ear height, ear width, number of rows per ear, number of kernels per row, number of kernels per ear and thousand kernel weights were also measured during the experiment. These are important traits which were frequently used to study drought and heat stresses in maize by Yousaf et al. (2022). The one of the most important trait i.e., kernel yield per hectare is measured according to the formula given and used by Tandzi & Mutengwa (2022).

$$\mathrm{G}\mathrm{r}\mathrm{a}\mathrm{i}\mathrm{n}\mathrm{Y}\mathrm{i}\mathrm{e}\mathrm{l}\mathrm{d}\left(\mathrm{k}\mathrm{g}/\mathrm{h}\mathrm{a}\right)={\displaystyle \frac{\mathrm{F}\mathrm{r}\mathrm{e}\mathrm{s}\mathrm{h}\mathrm{e}\mathrm{a}\mathrm{r}\mathrm{w}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}\left(\mathrm{k}\mathrm{g}/\mathrm{p}\mathrm{l}\mathrm{o}\mathrm{t}\right)\times \left(100-\mathrm{M}\mathrm{C}\right)\times 0.8}{\left(100-15\right)\times AreaHarvested/plot}\times 10,000}$$where;

MC = moisture contents of kernels.

Recording/measurement of physiological parameters and proline contents

Physiological parameters were analyzed from five, healthy and intact cob leaves per entry per treatment after impositions of drought stress treatments. The performance of maize hybrids under stress treatments was measured for three key physiological traits including net photosynthetic rate, stomatal conductance and transpiration rate using Infra-red gas analyzer (IRGA). The instrument used for the measurement was a handheld photosynthesis system (CI-320; CID Bio-Science Inc., Camas, WA, USA) as suggested and used by Yousaf et al. (2021). Moreover, the quantification of photosynthetic pigments i.e., chlorophyll a, b and carotenoid contents was done through spectroscopy using Shimadzu UV-1280 UV-VIS spectrophotometer (Peng & Liu, 1992; Lichtenthaler, 1987). The proline contents from maize leaves was quantified according to the procedure used by Efeoglu, Ekmekçi & Çiçek (2009).

Accumulation of reactive oxidative species (MDA and H₂O₂)

The estimation or quantification of MDA through thiobarbituric (TBA) method given and used by Heath & Packer (1968) was used in this study. Similarly, H₂O₂ contents were determined according to the procedure developed by Alexieva et al. (2001). The reaction mixture consisting of leaf extract, 2.5 mM potassium-phosphate buffer (pH 7.0) and 0.5 M potassium iodide (KI). After keeping the mixture in the dark for 1 h, the amount of H₂O₂ was determined through a spectrophotometer at 415 nm by reference to a standard curve prepared with H₂O₂ solutions.

Determination of enzymatic antioxidants (CAT and T-SOD)

The enzymatic activity of catalase (CAT) was measured by the method given by Chance & Maehly (1995) with a few modifications. After adding enzyme extract to start the reaction, its absorbance was measured every 20 s for 2 min at 240 nm. Each enzyme’s activity was expressed in terms of protein weight.

Measurement of kernel quality traits

Kernel quality parameters, such as the percentages of kernel protein, oil, and starch, were assessed using near-infrared spectroscopy (NIR) with the use of an Inframatic 9200 from Partin Instruments in Sweden. For each replication, an average of three samples (750 g each) was utilised to calculate the kernel quality attributes’ contents. Percentages (%) were used to express the parameters.

Statistical analysis

The recorded data for kernel yield and stress-associated traits were subjected to analysis of variance (ANOVA) and correlation coefficient analysis as suggested by Steel & Torrie (1997) through Statistix 8.1 (Analytical Software, Tallahassee, FL, USA) and XLSTAT (Addinsoft Inc, New York, NY, USA) statistical packages. The statistical differences between treatments were executed through Duncan’s multiple range test (DMRT). Biplot analysis was used to characterize maize hybrids concerning plant traits under different stress treatments by using XLSTAT 2020. Furthermore, OriginPro 2021 (OriginLab Corporation, Northampton, MA, USA) and Microsoft Excel 2019 were used for the graphical representation of data.

Metrological data (spring 2019–2020 and 2020–2021)

During the whole experimental study, data regarding metrological parameters i.e., minimum temperature/maximum temperature (°C), rainfall (mm), and relative humidity were recorded daily (Fig. 2). The data showed that the average daily maximum temperature was highest in May (40.6 °C and 38.9 °C) while the average daily minimum temperature was lowest in February (9.1 °C and 10.6 °C), during the spring growing season of 2019–2020 and 2020–2021, respectively. The rainfall data is not needed because the research area, where the experiment was conducted had the facility to cover the whole experimental area with a thick plastic sheet when needed. Indeed, the facility was used three times during the experiment to keep the rainfall away from the field.

Kernel yield and related traits

The results derived from analysis of variance (ANOVA) revealed that a significant difference was present among the maize hybrids for maize kernel yield and its associated morpho-phenological traits i.e., days to 50% anthesis (DA), days to 50% silking (DS), plant height (PH), ear height (EH), ear width (EW), rows per ear (R/E), number of kernels per row (NK/R), number of kernels per ear (NK/E), thousand kernel weight (TKW) and kernel yield (KY) under different drought stress conditions (Table 1). Moreover, the interaction between drought stress treatments and maize hybrids was also significant for all the studied traits.

The water stress at three critical plant development stages (pre-anthesis, anthesis, and post-anthesis) affected the overall development of the whole reproductive phase, which is, in fact, the most vulnerable stage to drought stress, causing a significant reduction in kernel yield and related traits (Table 2). Compared to control, water stress at the anthesis stage was proved more lethal than pre-anthesis and grain development stages as a significant reduction in most of the agro-morphological traits including days to 50% anthesis, days to 50% silking, ear width, rows per ear, number of kernels per row, number of kernels per ear, thousand kernel weight and kernel yield (Table 2). After the anthesis, water stress at pre-anthesis was the second most critical stage, severely affecting the plant and ear height of maize hybrids. However, ear length was harshly affected by the water stress at the grain development stage. Considering the performance of maize hybrids for morpho-phonological traits under control and water stress conditions at pre-anthesis, anthesis, and grain development stages, YH-5395 and YH-5427 outperform other maize hybrids i.e., YH-5482, NK-8441, and JPL-1908.

The correlation coefficient analysis revealed a significantly positive correlation of kernel yield with EL (0.77**), R/E (0.64**), NK/E (0.45*), EW (0.44*), and NK/R (0.33*) while significantly negative with PH (−0.79**), under control (Figs. 3A–3D). However, this correlation between morpho-phonological traits got changed in water stress conditions. There was a negative but non-significant correlation of KY with DT (−0.25^ns) and DS (−0.26^ns) under water stress at the anthesis stage, which was positive yet non-significant (0.030^ns, 0.017^ns) under control, respectively (Figs. 3A–3D). The correlation between the same traits under water stress conditions at the grain development stage was also positive (0.24^ns, 0.26^ns) but significant only for the anthesis stage (0.41*, 0.41*), respectively (Figs. 3A–3D). Similarly, the correlation of GY with EW, R/E, and NK/E were positive under control (0.44*, 0.64**, 0.45*), anthesis (0.21^ns, 0.34^ns, 0.18^ns), and grain development stage (0.40*, 0.30^ns, 0.35^ns), while negative under pre-anthesis water stress condition (−0.17^ns, −0.25^ns, −0.35^ns). The correlation of KY with PH was significantly positive under only the pre-anthesis water stress condition (0.43*) while negative under the control (−0.79**), anthesis (−0.17^ns), and grain development stages (−0.69*), respectively (Figs. 3A–3D).

Physiological traits and photosynthetic pigments

The results obtained from ANOVA unveiled the presence of highly significant differences (P ≤ 0.05) among maize hybrids, stress treatments, and their interactions for physiological traits and photosynthetic pigments i.e., net photosynthetic rate (Pn), stomatal conductance (Ci), transpirational rate (Tr), chlorophyll a and b (Chl a, Chl b) and carotenoid contents (carotenoid) (Table 1).

The impact of water stress on different plant physiological parameters at different growth stages was quite lethal, hindering the key physiological processes (Figs. 4A –4D and 5A–5D). The results revealed a significant reduction in major physiological traits and photosynthetic pigments i.e., net photosynthetic rate, stomatal conductance, transpirational rate, water use efficiency, chlorophyll a and b, and carotenoid contents under water-stress conditions during pre-anthesis, anthesis and grain development stages. Compared to control, water stress during the pre-anthesis stage was the most crucial period that resulted in the severe reduction of all the physiological traits and photosynthetic pigments. However, the impact of water stress during the anthesis phase was also at par with pre-anthesis water stress for physiological parameters. In maize hybrids, the net photosynthetic rate (Pn) was the most affected physiological trait under water stress conditions. Similarly, the negative effects of water stress on photosynthetic pigments in maize hybrids were significantly high except for carotenoid contents, which were less affected by water stress across the developmental stages. Leaf chlorophyll contents i.e., chlorophyll a and b (Chl a and Chl b) were significantly reduced under water stress conditions at the pre-anthesis stage (Figs. 5A–5D).

However, the impact of water stress during anthesis and grain development stages was statistically non-significant.

Considering the performance of maize hybrids under water stress conditions, two locally developed maize hybrids i.e., YH-5395 and YH-5427 appeared to be more drought tolerant than the other contending hybrids due to their relatively stable performance under stress conditions compared to other hybrids. However, their performance regarding the stability of photosynthetic pigments under water stress conditions was at par with a famous multinational maize hybrid NK-8441 of Syngenta Seeds.

Correlation analysis showed a varying degree of association between kernel yield and physiological traits and photosynthetic pigments under water stress conditions at different growth stages (Figs. 4A –4D and 5A–5D). The data revealed that kernel yield had a significantly positive correlation with net photosynthetic rate, stomatal conductance, transpirational rate, and chlorophyll-a under all water stress conditions. However, the correlation between these traits was a relatively quiet week at the pre-anthesis stage. Moreover, water use efficiency had a significantly positive correlation with kernel yield under control (0.74**) and anthesis water stress stage (0.88**), while negative under water stress at pre-anthesis (−0.079^ns) and grain development stage (−0.45*), respectively (Figs. 3A –3D). Similarly, chlorophyll b was positively correlated with kernel yield under control (0.88**) and water stress treatment at anthesis (0.59**) and grain development (0.30^ns) stages while negatively correlated at pre-anthesis water stress conditions (−0.048^ns) (Figs. 3A –3D).

Free proline contents

The analysis of variance showed the presence of significant genetic variation among maize hybrids for free proline contents under water stress conditions (Table 1). The results revealed that free proline contents were increased under water stress conditions in maize hybrids (Fig. 5D). Compared to control, the minimum effects of water stress were observed in the pre-anthesis stage, while the maximum increase of free proline contents was recorded in the grain development stage, indicating the susceptibility of the grain development stage to water stress. The differences between two water stress treatments i.e., anthesis and grain development were statistically non-significant. The correlation analysis also unveiled a strong positive correlation of free proline contents with kernel yield under control (0.88**) and water stress at different plant growth stages i.e., pre-anthesis (0.43*), anthesis (0.55*), and grain development (0.88**), respectively (Figs. 3A –3D).

Accumulation of reactive oxygen species (ROXs)

An increase in the accumulation of reactive oxygen species (ROXs) is one of the key responses of maize crops under osmotic/oxidative stress. Analysis of variance revealed significant differences among maize hybrids, stress treatments and their interactions for accumulation of ROS i.e., hydrogen peroxide (H₂O₂) and malondialdehyde (MDA) (Table 1). However, the variations were non-significant concerning years and their interactions with other sources of variations. Further results indicated a significant increase in H₂O₂ and MDA accumulation under water stress conditions (Figs. 6A –6D). Considering the accumulation of ROS under water stress conditions at different maize developmental stages, the highest accumulation of H₂O₂ and MDA was observed at the post-anthesis stage (post-anth) followed by the anthesis stage (during anthesis). The highest accumulation of H₂O₂ was recorded in NK-8441 and JPL-1908 (5.8 µmole g⁻¹ FW) followed by YH-5482 (5.4 µmole g⁻¹ FW) at post-anthesis stage while the lowest increase in ROS activity was observed in YH-5427 (5.1 µmole g⁻¹ FW). Similarly, the highest accumulation of MDA was recorded in NK-8441 (169.0 nmol g⁻¹ FW) followed (166.5 nmol g⁻¹ FW) at the post-anthesis stage while the lowest accumulation in YH-5427 (152.5 nmol g⁻¹ FW) at the pre-anthesis stage, respectively (Figs. 6A –6D). The correlation analysis unveiled a strong negative correlation of kernel yield with H₂O₂ and MDA under water stress at anthesis (r = −0.83**, r = −0.5*) and post-anthesis (r = −0.94**, r = −0.89 **), respectively (Figs. 3A–3D).

Accumulation of enzymatic antioxidants

With an increase in the production of reactive oxygen species (ROXs), plant triggers several mechanisms to protect cell components. Among these mechanisms, the production of antioxidants is one the most important mechanisms to scavenge and detoxify ROS species. In the current study, significant differences among maize hybrids, stress treatments and their interactions for the accumulation of antioxidants i.e., superoxide dismutase (SOD) and catalase (CAT) were found through analysis of variance (Table 1). However, the variations were non-significant concerning years and their interactions with other sources of variations.

Further results indicated a significant increase in SOD and CAT accumulation under water stress conditions (Figs. 6A –6D). The highest accumulation of SOD was observed during the water stress at the pre-anthesis stage (pre-anth) while for CAT, the highest activity was recorded during water stress at pre-anthesis and post-anthesis, respectively (Figs. 6A –6D). Concerning the performance of maize hybrids regarding the activity of antioxidants, the highest SOD and CAT accumulation was observed in YH-5395 (61.5 U mg⁻¹ pro, 60.5 U mg⁻¹ pro) and YH-5427 (8.6 U mg⁻¹ pro, 8.6 U mg⁻¹ pro). The correlation analysis unveiled a strong positive correlation of kernel yield with SOD and CAT under water stress at pre-anthesis (r = 0.92**, r = 0.27^ns), during anthesis (r = 0.90**, r = 0.92**) and post-anthesis (r = 0.96**, r = 0.70**) stages, respectively (Figs. 3A–3D).

Kernel quality traits

Analysis of variance revealed significant variations among maize genotypes for kernel quality traits across the treatments (Table 1). The results unveiled the significantly damaging effects of water stress (WS) on kernel quality traits at different maize development stages (Figs. 7A–7D). A significant reduction in kernel protein content percentage was observed in maize hybrids, especially NK-8441, where protein % decreased from 14.0% (Control) to 11.7% (WS at post-anthesis) (Figs. 7A–7D).

Among the crop development stages, water stress at the post-anthesis stage proved lethal for kernel protein content percentage. Similarly, kernel oil contents percentage was also highly affected by the water stress conditions at different stages (Figs. 7A–7D). The maximum oil % was observed in NK-8441 (5.4% at the control) while YH-5395 gave the minimum oil % (3.9%). The maximum decline in oil % was reported in YH-5482 and YH-5395 (25%) (Figs. 7A–7D). The impact of water stress at different plant stages for oil % was random. Water stress also affects the kernel starch content percentage which is another very important maize kernel quality parameter. The highest starch % was observed in exotic, multinational hybrid NK-8441 (53.3%) at control while the lowest in the JPL-1908 (44.7%) during water stress at the pre-anthesis stage (Figs. 7A–7D). The maximum reduction in starch percentage (10.8%) was observed in maize hybrid YH-5427, from 49.9% in control to 45.0% at pre-anthesis water stress (Figs. 7A–7D). The highest impact of water stress was recorded in the pre-anthesis stage, where reduction in starch percentage ranged from 3% to 10.8%.