Selenium Application Improves Drought Tolerance during Reproductive Phase of Rice

Abstract

Rice (Oryza sativa L.) grain yield is reduced by drought stress during reproductive stages. In this work, the effects of foliar selenium (Se) spray on the reduction of drought stress are measured and the most critical stage of rice is identified. Three field experiments were carried out. In experiments I and II, the sensitive stage was determined by imposing drought stress at the panicle initiation and flowering stages by withholding irrigation for 10, 15, 20, and 25 days. In experiment III, drought stress was induced by depriving plants of moisture for 20 or 25 days. Se spray was then applied at rates of 0, 10, and 20 mg L⁻¹ to alleviate the drought, and its effects were evaluated. Overall, it was reported that the panicle initiation stage was more vulnerable to drought stress than the flowering stage, as evidenced by a larger reduction in grain yield. Employing Se as a foliar spray at a rate of 20 mg L⁻¹ under drought stress enhanced the number of filled grains m⁻², the total number of grains m⁻², and grain yield ha⁻¹ by 22.0, 4.3, and 11.0%, respectively, over water spray. This shows a positive interaction or synergistic interaction between drought tolerance and selenium application. These results demonstrate that foliage supplementation of Se at a dosage of 20 mg L⁻¹ alleviate the adverse effects of drought stress at the panicle initiation stage in rice cultivation. Se mitigated the negative effects of drought by improving the membrane stability index, relative water content, and proline content.

1. Introduction

Rice is the most extensively consumed staple food in the world, offering 540 kcal of energy and accounting for more than half of all human intake, as well as being abundant in nutrients, vitamins, and minerals. The average rice output is anticipated to be 5.0 × 108 t, with the demand expected to rise to 2.0 × 109 t by 2030 due to population growth [1]. The current and expected global situation necessitates a significant increase in rice output under less favorable rainfed soils. Increasing air temperatures, variability in rainfall, and an increase in the frequency of extreme weather events related with temperatures and water stress are major global threats to agricultural output [2]. Drought is a significant abiotic factor that limits rice productivity in the rainfed or irrigated rice ecosystems [3]. Global climate change models forecast altered patterns of precipitation and elevated air temperatures, which lead to arid conditions during the growing season of many crops [4].

Drought-related rice production losses are expected to grow by about 19% across Asia by the end of the twenty-first century if no adaptation strategies are implemented [5]. Furthermore, such yield losses due to drought will be more likely in regions with less developed irrigation facilities, as rice production requires more water than other crops [6]. The severity of drought stress varies and greatly depending on rainfall frequency, soil moisture content, and evaporation [7]. Drought stress reduces turgor pressure, leaf water potential, and stomatal conductance in plants, resulting in decreased cell development and enlargement [8]. Drought stress also reduced chlorophyll content, photosynthesis, respiration, carbohydrate metabolism, nutrient uptake, and translocation [8,9]. Rice productivity will be reduced by 200–600 kg ha⁻¹ due to variations in drought intensity and frequency (up to 20%) by 2050 [10]. Therefore, it is evident that drought stress can decrease the rice grain yield, and any crop management options to sustain the crop yield under drought stress can increase rice productivity.

Selenium (Se) is a trace mineral necessary for various body functions. Recommended Se intake for adults is 25–34 µg day⁻¹, while 6–22 µg day⁻¹ for children. Se is a potent antioxidant that helps the human body to combat oxidative stress and protects from chronic illnesses such as heart disease and cancer. Similarly, plants often act as antioxidants at lower concentrations and protect from drought and high-temperature stress [11,12]. Low dosages of Se have protected wheat (Triticum aestivum L.) seedlings against cold stress [13], sorghum from high temperature effect, rapeseed (Brassica napus L.) seedlings from drought [14], and silver maple (Acer saccharinum L.) from desiccation [15]. Under drought stress conditions, Se at 3 mg L⁻¹ enhanced the number of branches in okra (Abelmoschus esculentus L.) [16]. Se increased membrane stability index and relative water content in strawberries (Fragaria × ananassa) [17]. During drought conditions, foliar spray of Se caused proline accumulation in sesame (Sesamum indicum L.) leaves [18], and soil application increased the peroxidase enzyme activity [19]. Wheat plants fertilized with Se have increased yield under drought stress by increasing the number of productive tillers, number of grains spike⁻¹, and grain yield [20]. Apart from bulk Se, nano-Se also improved crop abiotic stress tolerance. For example, the application of 30 mg L⁻¹ Se nanoparticles to wheat seedlings during drought stress has improved plant height and shoot length [21], and in maize, nano-Se (2.5 mg kg⁻¹) caused a significant increase (17.89%) in plant height [19]. Furthermore, exogenous Se treatment is useful in combating water stress in a variety of crops, including canola [22], soybean [23], quinoa [24], and edamame [25]. As a result, it is evident that foliar application of Se can mitigate the negative consequences of drought stress. To the best of our knowledge, there is no evidence that Se can mitigate the negative effects of early reproductive stage drought stress on rice. Our present research mainly caters the need of bringing out an alternative option to mitigate drought stress in rice cultivation. The emerging circumstances of increased global temperature and increased frequency of weather anomalies will throw a challenge of reduced water availability for crop production. Identification of some risk-reducing agronomic interventions to mitigate drought in rice cultivation could be feasible for adoption at farmer’s scale. Hence, we hypothesize that the foliar spray of Se can improve the rice grain yield by its inherent antioxidant activity under drought conditions. The main objectives of the experiments were to identify whether individual or combined growth stages, viz., panicle initiation and flowering stage, are susceptible to drought stress and to investigate the effect of Se spray at that critical stage on rice yield.

2. Materials and Methods

2.1. Experiment I and II

2.1.1. Plant Material and Growing Conditions

During the Kharif season, two field experiments were carried out at the Tamil Nadu Agricultural University’s wetland farm (11° N latitude and 77° E longitude, altitude 426.7 m above MSL), Coimbatore, Tamil Nadu, India (July–November 2019). During Experiment I, the total rainfall received was 116.1 mm, while the average maximum and minimum temperatures were 30.6 °C and 23.0 °C, respectively. Rainfall occurred following the drought stress period. Rice CO-51, a short duration (110 d) cultivar, was deployed as a test variety. The available N status in the soil of the experimental plot was low (213.2 kg ha⁻¹), the available P₂O₅ was high (47.2 kg ha⁻¹), and the available K₂O was also high (642.8 kg ha⁻¹). The average maximum and lowest temperatures were 31.6 °C and 21.9 °C, respectively, with a total rainfall of 65.1 mm. In the experimental plot, the soil reaction (pH), EC (1:2.5 soil-water suspension dS m⁻¹), and organic carbon (g kg⁻¹) were all 7.8, 0.37, and 6.7, respectively. In the experimental plot, the soil had low available nitrogen (238.0 kg ha⁻¹), high available phosphorus (42.0 kg ha⁻¹), and high available potassium (1046.8 kg ha⁻¹). The soil texture of the experimental site was clay loam in both the experiment I and II with average sand, silt and, clay percentage of 42.0%, 11.6%, and 46.2%, respectively.

2.1.2. Crop Husbandry and Treatment Details

Pre-germinated seeds (7.5 kg ha⁻¹) were sown evenly in a nursery area of 15 m², which were pre-soaked overnight and incubated for 24 h. After thorough puddling, the field was uniformly levelled with the help of a wooden board, and the bunds were made. The gross plot size was 8.25 m² (5.5 m × 1.5 m). Fourteen-day-old seedlings were carefully removed from the nursery and transplanted in the main field. Single seedling hill⁻¹ was planted at a spacing of 25 × 25 cm. The soil was fertilized with 150:50:50 kg N:P₂O₅:K₂O, using urea (46% N), single superphosphate (16% P₂O₅), and muriate of potash (60% K₂O), respectively, as a source of fertilizer. Nitrogen and potassium were applied in four equal splits viz., at basal, active tillering (35 DAS), panicle initiation (50 DAS), and flowering (70 DAS) stages. Phosphorus (100%) was applied as basal. Cono-weeding was carried out at 15, 25, 35, and 45 DAT to control weeds. For the irrigated control treatment, 5 cm of standing water was maintained in the field from transplanting to maturity. Under irrigated-control (no stress) treatment, irrigation was provided whenever hairline cracks were developed (average moisture content was ranged from 32.9 to 34.8%, dry weight basis), as described in the System of Rice Intensification (SRI). Chlorpyriphos (2 mL L⁻¹) was sprayed to control the yellow stem borer, and malathion (1 mL L⁻¹ of water) was sprayed to control the ear head bug at the recommended spray dose. In both control and drought stress treatment, irrigation was stopped 10 d before harvest and harvested at physiological maturity. The grain and straw were separated using a mechanical thresher. The dates of sowing and harvest dates were 3rd July 2019 and 8th November 2019 in Experiment I. The experiment was repeated once with the same treatments and crop husbandry, as explained in experiment I. The second crop was sown and harvested on 4th December 2019 and 30th March 2020, respectively.

2.1.3. Drought Stress Imposition in the Field

Treatment Details

Three replications of the split-plot design were used for the experiment. The duration of the drought stress is the subplot, while the growth stage is the main plot. The panicle initiation stage and the flowering stage were the two stages of the main plot. Five levels of drought stress for 0, 10, 15, 20, and 25 days made up the subplot. A polythene sheet of 30 cm long was placed in the ground to a depth of 20 cm along the bunds of each plot to restrict seepage of water from one to the other. Buffer channels of 1 m were also maintained between the irrigated and drought-stressed plots. The subplot, namely drought stress, was imposed by withholding irrigation at panicle initiation and flowering stages as per the treatment schedule. Irrigation was provided after the concerned treatment period was over. The panicle initiation stage was confirmed by dissection and visualization under a microscope. For identification of initiation of panicle, mother tiller was dissected longitudinally into half and visualized under microscope. A furry tip was seen at the base of the tiller under microscope which indicated initiation of panicle. After identifying the stage, the water was withheld as per the treatment schedule.

2.1.4. Parameters Recorded

A one-meter wooden scale was used to measure the height of five labelled plants. Plant height was measured in centimetres from the base of the plant to the tip of the longest leaf during the panicle initiation stage and from the base of the plant to the tip of the panicle during the blooming period.

At physiological maturity, the number of tillers produced in the five tagged hills was recorded.

Relative water content was calculated as explained by [26] and expressed in %.

where, Fw—Fresh weight of leaf sample (mg), Tw—Turgid weight of leaf sample (mg), and Dw—Dry weight of leaf sample (mg).

Fresh leaves from the fully grown third leaf were taken from each treatment. After removing the top and bottom portions of the leaf, a 5 cm section was cut from the leaf central portion and kept in a pre-weighed empty centrifuge tube with the cap air tightened. The samples were brought to the lab, where they were weighed for the first time (Fw). The centrifuge tubes were filled with distilled water and sealed with a lid to achieve turgid weight. After 8 h, the leaf was gently removed, cleaned with filter paper to eliminate water droplets that had adhered to the leaf surface, and the turgid weight was measured (Tw). The leaf samples were then set in a butter paper cover and processed for 48 h at 60 ± 5 °C in a hot air oven and dry weight was noted.

At the panicle initiation and flowering stages, cell membrane stability was assessed as per [27]. After placing 0.1 g of leaf bits in a test tube, 10 mL of distilled water was added, the solution was kept for half an hour, and the initial EC was determined after the leaf bits were removed. The leaf bits were then immersed in the same solution and boiled for 10 min in a water bath at 100 °C. After the timer ran out, the solution was cooled, the leaf bits were removed, and the final EC of the solution was recorded.

Membrane stability index = [1 − (C1/C2)] × 100, where C1—Electrical conductivity was measured at 40 °C test tubes, and C2—Electrical conductivity was measured at 100 °C test tubes.

The chlorophyll stability index was calculated using the method provided by [28]. 0.5 g of leaf sample was added to a set of test tubes having 20 mL of water and maintained at 65 °C for 30 min in water bath. The material was then macerated with 10 mL of 80% acetone, centrifuged at 3000 rpm for 10 min, and the volume was made to 25 mL. To quantify the total chlorophyll content, the optical density was measured at 652 nm and expressed as a percentage.

Proline was estimated as described by [29]. Leaf samples were collected from the field using airtight sampling tubes. Proline was isolated from the third leaf from the top by macerating the leaf sample with 3% sulphosalicyclic acid and centrifuging it at 3000 rpm for 10 min. Following extraction, 1 mL of supernatant was added to a test tube, followed by 2 mL of acid ninhydrin solution, 2 mL of 6 M orthophosphoric acid, and 2 mL of glacial acetic acid. The content was boiled in a hot water bath for 30 min at 50–60 °C till the pink color developed. The solution was then cooled to room temperature before being separated using a separating funnel using 4 mL of toluene. At 520 nm, the optical density was measured. As a standard, pure proline was employed and represented as μg g⁻¹.

Productive tillers’ hill⁻¹ number was recorded in five random hills in each plot, the average was taken and expressed in numbers m⁻². The tagged plants were harvested with care, and ten panicles were selected, hand threshed, and the average grains panicle⁻¹ were counted. Ten panicles were selected from five tagged hills, and the grains in each panicle were divided into filled grains and chaffy grains. The number of panicle⁻¹ filled grains was counted and reported as a number. Harvested plants from each net plot area were threshed, washed, and dried before being weighed in kg ha⁻¹ at a moisture level of 14%. Straw was sun-dried from each net plot area and the weight was measured in kg ha⁻¹.

2.2. Experiment III—Effect of Drought and Se Spray on Rice Growth, Physiology, and Yield

2.2.1. Growing Environment Characterization

A field experiment was carried out at a wetland farm maintained by Tamil Nadu Agricultural University in Tamil Nadu, India (11° N latitude, 77° E longitude, altitude of 426.7 m above MSL). The crop was planted on 24 February 2021, and it was harvested on 18 June 2021. The total amount of rain that fell during the cropping season was 71.5 mm. The mean maximum and lowest temperatures were, respectively, 34.2 °C and 23.8 °C. There was no rain throughout the drought stress period. The soil’s texture was clay loam, and it had an EC of 0.44 dS m⁻¹ and a pH of 8.1. The soil had an organic carbon content of 5.5 g kg⁻¹. Low available nitrogen (224.2 kg ha⁻¹) and high available phosphorus (33.4 kg ha⁻¹) and potassium (618.3 kg ha⁻¹) made up the soil’s nutritional status. With a composition of 45.6% clay, 11.8% silt, and 42.8% sand, the soil is classified as a clay loam.

2.2.2. Treatment Details

The experiment was laid out in a split-plot design with three replications. The CO-51 was used in this experiment. The main plot was irrigation regime, and the subplot was foliar spray of selenium. The gross plot size was 5.0 m × 4.0 m. Main plot treatments were M₁—Irrigation as in the System of Rice Intensification (SRI) method; M₂—Drought stress for 20 d from panicle initiation stage; and M₃—Drought stress for 25 d from panicle initiation stage and subplot treatments were S₁—water spray; S₂—Se foliar spray at 10 mg L⁻¹; and S₃—Se foliar spray at 20 mg L⁻¹. The duration of drought stress was fixed from experiment I and II and the concentration of Se spray was based on earlier experiments.

2.2.3. Crop Husbandry

The crop husbandry practices followed in Experiment—III were similar to Experiment—I and Experiment—II.

2.2.4. Treatment Imposition

Drought stress was imposed at PI stage for 20 d and 25 d, similar to experiments I and II. Drought stress was relieved by providing irrigation as per the treatment schedule. Sodium selenate (Na₂SeO₄) was employed as a Se source. Weighed 1.0 g of sodium selenate and diluted it in 1000 mL of water in a 1.0 L volumetric flask to make 1000 mg L⁻¹ Se stock solution. 10 mL and 20 mL of the stock solution were taken and made into 1000 mL to obtain 10 mg L⁻¹ and 20 mg L⁻¹ of Se concentrations. A 500 L ha⁻¹ spray volume was employed. For each plot (20 m²), 1.0 L of Se was sprayed using a tiny hand sprayer. Water was sprayed in the control plots. Two foliar sprays were given after drought stress treatment in each plot in a one-day interval.

2.2.5. Parameters Recorded

Similar to experiment I and II, growth, physiological and yield parameters were recorded.

2.3. Data Analysis

Three replications of a split-plot design were used for all of the trials. The SAS software’s PROC MIXED procedure was used to analyze the data [30]. Initial data analysis was carried both separately and combinedly for each experiment (Experiment I and Experiment II). Similar responses and significance levels were found for all traits in the results of all tests, whether conducted separately or together. There were also no interaction effects. As a result, the next chapter presented explains the mean responses from two experiments (each with three replications and six total). LSD was used to differentiate between the means.

3. Results

3.1. Experiment I and II

3.1.1. Morphological Traits

There was a significant difference (p < 0.05) for plant height and number of tillers per m⁻² due to irrigation regime, duration of drought stress and their interaction (Figure 1). Drought stress for 25 d at PI and flowering stages reduced the plant height by 24.4 and 4.9%, respectively. The tiller number per m⁻² was decreased by 42.7 and 29.3% due to drought for 25 d at PI and flowering stages, respectively, compared to irrigated control.

3.1.2. Physiological and Biochemical Traits

The RWC (%), MSI (%), CSI (%) and proline (µg g⁻¹) significantly varied (p < 0.05) for irrigation regime, duration of drought stress and their interaction (Figure 2 and Figure 3). The RWC was reduced by 43.2 and 29.7% due to the drought period of 25 d at PI and flowering stages, respectively. The reduction of CSI due to drought stress of 25 d at PI and flowering stages was 16.4 and 19.1%, respectively compared to irrigated control. There was a reduction in MSI (%) due to drought stress for 25 d at PI and flowering stages by 20.2 and 30.9%, respectively, compared to irrigated control. The proline content was increased by 528 and 699% due to drought stress at PI and flowering stages, respectively, compared to irrigated control.

3.1.3. Yield and Yield Contributing Characteristics

The effect of irrigation regime, duration of drought stress and their interactions were significant (p < 0.05) for grain yield, straw yield, total grains panicle⁻¹ and filled grains panicle⁻¹ (Figure 4 and Figure 5). The pooled data for two seasons indicate that there was a 57.9 and 42.6% yield penalty due to drought at PI and flowering stages, respectively, compared to irrigated control. The straw yield was reduced by 28.7 and 36.3% for 25 d of stress at PI and 25 d of stress at the flowering stages, respectively, over irrigated control.

The total grains and the filled grains panicle⁻¹ varied significantly (p < 0.05) due to 25 d of drought stress each at PI and flowering stages (Figure 5). The pooled data of two seasons show a reduction in the total grains’ panicle⁻¹ due to drought stress for 25 d at PI and flowering stage by 25.3 and 7.7%, respectively when compared to irrigated control. Similarly, the total filled grains panicle⁻¹ was reduced by 31.2 and 16.1% due to drought stress at PI and the flowering stage for 25 d, respectively (Figure 5).

3.2. Experiment III

3.2.1. Main Effect of Drought Stress

Morphological Traits

Due to drought stress, there was a significant difference in plant height and the number of tillers at the panicle initiation stage (

Table 1. Main effect of irrigation regime on morphological, physiological, and yield and yield associated traits of rice grown under puddled conditions.

Trait	Irrigation Regime			LSD
	Irrigated Control	20 Days Drought Stress	25 Days Drought Stress
Morphological traits
Plant height (cm)	99.3 ± 0.92 a	91.4 ± 0.92 b	90.6 ± 0.92 b	2.8
Number of tillers m−2	544.3 ± 13.1 a	458.3 ± 13.1 b	390 ± 13.1 c	39.2
Physiological traits
Relative water content (%)	87.7 ± 0.44 a	67.8 ± 0.44 b	64.3 ± 0.44 c	1.3
Proline content (µg g−1)	43.7 ± 8.56 b	397.7 ± 8.56 a	404.9 ± 8.56 a	25.4
Chlorophyll stability index (%)	82.4 ± 0.57 a	68.0 ± 0.57 b	64.7 ± 0.57 c	1.7
Membrane stability index (%)	81.8 ± 0.35 a	69.4 ± 0.35 b	63.4 ± 0.35 c	1.04
Yield and yield associated traits
Number of filled grains m−2	182.3 ± 1.76 a	153.8 ± 1.76 b	142.6 ± 1.76 c	5.2
Total grains m−2	204.6 ± 3.0 a	181.3 ± 3.0 b	176.8 ± 3.0 c	8.9
Grain yield (kg ha−1)	6909 ± 51 a	6195 ± 51 b	5948 ± 51 c	151.9
Straw yield (kg ha−1)	9659 ± 124 a	7158 ± 124 b	7139 ± 124 b	370.6

Note: Letters a, b, c indicate that grouping of means is done based on LSD.

). Drought stress for 20 and 25 d reduced plant height by 7.9, and 8.8%, respectively and the number of tillers was decreased by 15.8 and 28.3% compared to control.

Physiological Traits

The proline content (µg g⁻¹), RWC (%), MSI (%), CSI (%) were significantly varied among treatments (Table 1). Drought stress for 25 d decreased RWC, CSI, and MSI by 26.6, 21.4 and 22.4%, respectively whereas drought stress for 20 d decreased them by 22.6, 17.5, and 15.1%, respectively, compared to control. A 25-day drought stress elevated proline content by 826%; however, drought stress for 20 d increased by 810% compared to irrigated control.

Yield and Yield Associated Traits

The number of filled grains m⁻², grain yield (kg ha⁻¹), and straw yield (kg ha⁻¹) was significantly reduced by 15.6, 11.4, 10.3, and 25.9%, respectively owing to drought stress for 20 d. on the other hand, drought stress for 25 d reduced it by 21.8, 13.6, 13.9, and 26.1%, respectively compared to control (Table 1).

3.2.2. Main Effect of Foliar Spray of Se

Morphological Traits

Plant height and the number of tillers m⁻² were significantly increased by foliar spray of Se at the rate of 20 mg L⁻¹ by 5.4 and 12.6%, respectively, compared to water sprayed plants (

Table 2. Main effect of selenium foliar spray on morphological, physiological, and yield and yield associated traits of rice grown under puddled conditions.

Trait	Foliar Spray			LSD
	Water	Se @ 10 mg L−1	Se @ 20 mg L−1
Morphological traits
Plant height (cm)	91.0 ± 0.92 b	94.5 ± 0.92 a	95.9 ± 0.92 a	2.8
Number of tillers m−2	430.9 ± 13.1 b	476.4 ± 13.1 a	485.3 ± 13.1 a	39.2
Physiological traits
Relative water content (%)	67.5 ± 0.44 c	75.3 ± 0.44 b	76.9 ± 0.44 a	1.3
Proline content (µg g−1)	251.1 ± 8.56 b	290.8 ± 8.56 a	304.3 ± 8.56 a	25.4
Chlorophyll stability index (%)	64.1 ± 0.57 c	73.9 ± 0.57 b	77.1 ± 0.57 a	1.7
Membrane stability index (%)	66.0 ± 0.35 c	73.5 ± 0.35 b	75.3 ± 0.35 a	1.04
Yield and yield associated traits
Number of filled grains m−2	142.9 ± 1.76 c	161.8 ± 1.76 b	174.0 ± 1.76 a	5.2
Total grains m−2	183.4 ± 3.0 a	187.9 ± 3.0 a	191.3 ± 3.0 a	8.9
Grain yield (kg ha−1)	6032 ± 51 c	6323 ± 51 b	6698 ± 51 a	151.9
Straw yield (kg ha−1)	7524 ± 124 b	8144 ± 124 a	8287 ± 124 a	370.6

Note: Letters a, b, c indicate that grouping of means is done based on LSD.

).

Physiological Traits

Foliar Se spraying at a rate of 20 mg L⁻¹ markedly enhanced the RWC (%), proline content (µg g⁻¹), CSI (%), MSI (%) by 13.9, 17.5, 20.3 and 14.1%, respectively in comparison to control (Table 2).

Yield and Yield Associated Traits

The number of filled grains m⁻², total grains m⁻², grain yield (kg ha⁻¹), and straw yield (kg ha⁻¹) was significantly increased up to 21.8, 4.3, 11.0 and 10.1%, respectively by foliar application of Se at the rate of 20 mg L⁻¹ compared with water spray (Table 2).

4. Discussion

The intensity and frequency of drought occurrence have become unpredictable due to climate changes; Consequently, it is crucial to comprehend how plants respond to drought stress. Rice is vulnerable to drought stress at the panicle initiation, flowering, and grain filling phases, according to earlier research studies [31]. The plant height and number of tillers m⁻² were both decreased by drought stress at the initiation of the panicle for 25 days, which is greater than the flowering stage (Figure 1). The number of tillers m⁻² may be more significantly impacted by the timing of moisture stress. It is demonstrated by the fact that the number of tillers m⁻² decreased most dramatically during the early reproductive stage (PI stage) compared to the later reproductive stage (flowering), when the plants had finished the production of tillers.

This study showed a significant decrease in the RWC at the PI stage compared to the flowering stage (Figure 2). In contrast, the proline content was the highest during flowering stage drought stress. The higher decrease in the RWC at the PI stage could be attributed to the disruption of root hydraulic conductivity [32] and soil moisture depletion. Studies on cotton [33], wheat [34], and lentil [35] have shown the reduction in RWC during drought stress due to decreased cell turgor as observed in this study.

Drought stress increased the membrane damage. It was higher at the PI stage than in the flowering stage, which could be due to enhanced production of oxidants and lesser activity of antioxidant enzymes [36]. In maize and lentils, drought stress had increased the membrane damage [36,37]. The capacity to withstand drought is greater at the flowering stage than at the PI stage, as per higher proline accumulation at this period. This was well established by [38]. Our study showed a decrease in the CSI (16.4 and 19.1% at PI and the flowering stages, respectively), which could be due to decreased leaf RWC (Figure 3). Higher CSI at the flowering stage than the PI stage indicates that the PI stage is more sensitive than the flowering stage.

The total number of grains panicle⁻¹ and filled grains panicle⁻¹ were reduced severely (25.3 and 31.2%) at the PI stage compared to the flowering stage. Decreased grain numbers in rice and wheat were due to pollen and spikelet abortion when drought stress coincided with the early reproductive phase [39,40]. Furthermore, the role of leaf turgor is directly associated with the photosynthetic ability of the plants. Grain yield and relative water content were positively correlated in rice [41]. The substantial reduction in grain output (57.9%) was brought on by 25 days of drought stress during the PI stage as compared to the flowering stage (42.6%). Our research showed that rice is more drought-sensitive while it is developing gametes as opposed to when it is in blossom. This is consistent with the findings of several researchers [42,43]. Mild drought stress during the pre-anthesis stage resulted in a reduction of 45% of spikelet panicle⁻¹ and 70% of secondary branches [39]. The drop in leaf water status, which was caused by a decrease in sucrose production and its translocation to flowers, had an impact on the reproductive function at the PI stage [36,44]. High levels of abscisic acid driven by drought stress during the PI stage may prevent pollen from receiving enough sucrose, which might reduce pollen viability [45].

The association between Se application and water stress was non-significant since the foliar spray of selenium boosted growth and yield under both drought stress and irrigated conditions. As a whole, this was insignificant. Under drought stress, foliar application of Se increased plant height and number of tillers m⁻² more than water spray. Foliar application of Se resulted in enhanced RWC, CSI, MSI, and proline content. The previous study has shown that foliar spray of Se at lower concentrations improved the drought stress tolerance [46]. During drought conditions, exogenous application of Se increased water content in plant tissues, and it could be due to increased water uptake through a thick and active root system [47]. This study showed that Se improved the drought tolerance of rice by retaining more chlorophyll content and increased osmolyte accumulation. Se might decrease the formation of oxidants by increasing the antioxidant property [48,49]. Production of proline in Se applied plants might be because of its role in proline metabolizing enzymes. Se affects proline metabolism by changing glutamyl kinase and proline oxidase, resulting in increased proline synthesis and decreased breakdown. It was previously evidenced that foliar spraying of Se during drought stress caused proline accumulation in sesame (Sesamum indicum L.) [18]. Wheat seedlings treated with Se had more proline, which improved the water status of drought-stressed plants [47]. Foliar spraying of Se improved the number of filled grains, total grains plant-1, grain and straw production. Increased grain yield with Se treatment under severe drought might be attributed to antioxidant induction and up-regulation [50,51].

5. Conclusions

Our research concluded that drought stress at the PI stage decreased rice grain yield and yield-contributing attributes greater than it did during the flowering stage, indicating that the PI stage is more vulnerable to drought stress. As a result, we suggest using Se via foliar spray during panicle initiation stage in rice helps to mitigate the negative impacts of drought stress on paddy production. Selenium treatment indirectly confers drought resistance in rice plants by increasing water absorption, RWC, CSI, MSI, and proline content.