Development of Rice Mutants with Enhanced Resilience to Drought Stress and Their Evaluation by Lab Assay, Field, and Multivariate Analysis

Drought is one of the foremost devastating abiotic stresses reported for rice crops. To improve the productivity of rice, diversity is being enlarged by induced mutation using a source of gamma rays. But this type of mutation rarely results in fruitful products because the chances of getting the desired mutant are very low. The present study aimed to evaluate the rice mutants against drought or osmotic stress. In this study, three experiments were conducted that comprised of seventy-one mutants originating from different doses of gamma rays (Cs137) along with parent RICF-160 and commercial variety (Kainat) were tested. In the first experiment, germination and seedling attributes were calculated under control and osmotic stress conditions created by using 16% (0.6 MPa) polyethylene glycol (PEG-6000). Results revealed that all the mutants exhibited significant (p < 0.01) responses to PEG-induced osmotic stress. Principal component biplot analysis (PCBA) revealed the first seventeen cumulative PCs with eigenvalues >1 contributed 88%. It was noted that the germination percentage (GP), germination rate (GR), coefficient velocity of germination (CVG), and seed vigor (SV) contributed maximum and positively in PC1. Results showed the highest germination percentage (GP) at 48 hrs in mutant NMSF-11 (88.9%) followed by NMSf-38 (73.3%). Similarly, the germination rate (GR) and coefficient velocity of germination (CVG) were measured highest in NMSF-11 (9.7 and 118.1%), respectively. In stress conditions, the mutants NMSF-35 and NMSF-36 depicted the highest GP, GR, and CVG. The maximum seed vigor (SV), shoot length (SL), root length (RL), and fresh weight (FW) were observed in mutants NMSF-50 and NMSF-51 under both conditions, whereas the mutants NMSF-59, NMSF-60, NMSF-64, and NMSF-67 showed lower values for SV, SL, RL, and FW. In the second experiment, a field trial was conducted at the Nuclear Institute for Agriculture and Biology (NIAB), Faisalabad, in two control and stress sets. A bit different trend was observed among all mutants for agronomic parameters under both conditions. In the third experiment, biochemical profiling was done in Marker Assisted Breeding (MAB) Lab-1, Plant Breeding and Genetics Division. A significant variation was seen in enzymatic antioxidants and chlorophyll content in both control and stress conditions. Under control conditions, the ascorbate peroxidase (APX) content was observed higher in mutant NMSF-49 (106.07 Units/g. f. wt.). In comparison with the stress, the ascorbate peroxidase activity was higher in NMSF-41 (82.34 Units/g. f. wt.). Catalase (CAT) activity was observed maximum in NMSF-29 (17.54 Units/g. f. wt.) and NMSF-40 (14.17 Units/g. f. wt.) under control and stress conditions, respectively. Peroxidase (POD) activity was observed maximum in NMSF-51 (22.55 Units/g. f. wt. and 10.84 Units/g. f. wt.) under control and stress conditions, respectively. In conclusion, to fit in the changing climate conditions for resilient rice crop production, the promising mutant lines may be used to transfer the desirable drought-tolerant/drought-resistant genes in rice germplasm.


Introduction
South Asia is the most sensitive to climate change among various agroecologies and has a large population with limited access to essential resources, such as water and arable land [1].Most country priorities for improving drought tolerance in crops include increasing agricultural water productivity under rain-fed conditions [2,3].Plant growth and development are signifcantly hampered due to water shortage resulting in lower crop yields [4,5].Drought creates osmotic stress reactive oxygen species (ROS) in plant [6,7] and animal [8,9] cells that cause oxidative damage [10,11] and is one of the reasons for gaps in world food production [2,12,13].Drought tolerance can be achieved in two ways: by shortening the life cycle (escape) or by developing morphophysiological adaptations [14], such as root architecture, water use efciency, and molecular changes to withstand the stress condition (true tolerance) [2,[15][16][17].Drought obviously afects the crop productivity and hinders the seed germination [18][19][20][21].Its main afect is on the growth and cell expansion [22,23], which leads to a reduction in the shoot and root length in rice [5,24,25].
A recent estimate on climate change predicts that the intensity and frequency of drought will be of high magnitude in the future and will exhibit a signifcant impact on rice crop production [26,27].Te drought stresses created by low rain-fed conditions highlighted the future scenario to have rice varieties suitable for the water stress conditions that may sustain rice crop production under variable climate conditions [28,29].
To improve rice crops against water stresses, diferent mutation breeding techniques have been used successfully in rice.Te mutagenesis in rice is advantageous due to its small genome, i.e., a small population is required to saturate the whole genome and to provide a larger allelic series for use in mutagenesis [30,31].Random mutations caused by physical and chemical agents have been applied to create genetic variability and for gene functional studies in rice [32].In mutation breeding, instead of crossing, we just expose the high yielding cultivated variety to mutagen (physical and chemical) for improvement against one or other desirable genes without compromising the rice grain quality attributes [33].Tis approach takes less time to uniform the breeding material and is non-GMO in practice.Induced mutation is the briefest conceivable strategy for the advancement of Basmati rice varieties/germplasm [34,35].Mutation breeding through gamma rays is a powerful and exceptionally fruitful methodology for the generation of commercial cultivars [34,36].Te utilization of induced mutation in crop improvement has been demonstrated to be a successful way to deal with improved yield, quality, and protection from abiotic stresses [37,38].Induced mutation has been utilized in rice more than any other crop as confrmed by more than 443 rice mutant varieties listed in the FAO/IAEA Mutant Varieties Database [39,40].
Rice (Oryza sativa L.), a monocot with shallow root architecture yet demand for high water for growth, is more susceptible to water shortage stress than other main food crops [41].About 50% of the total populace (340 million individuals in South Asia and 140 million in Southeast Asia and sub-Saharan Africa) depends upon rice crop.Globally, Pakistan is known as one of the major rice-producing countries, where the area under rice crop is about 3.034 million hectares and contributes 3.1 percent of value added in agriculture and 0.6 percent in GDP (Pakistan Economic Survey, 2019 −20).
To accelerate the breeding programs, screening strategies for drought resistance must be quick, cost-efective, and consistent for evaluating plant performance at the seedling stage [34,36,42].Screening at seed and seedling stage is a reliable strategy and accelerates the process of screening stress-tolerant crops [43][44][45][46][47].So the genetic association between germination and seedling traits will further help the breeders to improve drought tolerance in rice.Cluster analysis classifes drought-tolerant and drought-susceptible rice mutants for desirable traits based on genetic similarity.Tus, there is need to identify the suitable selection criteria that will enhance the efciency of breeders to select droughttolerant mutants from diverse populations.Te current study was designed to evaluate diverse rice mutants developed through induced mutation, which have a higher tolerance to drought or osmotic stress in both lab and feld conditions.
Te analysis of variance of means showed that the variation among the mutants was highly signifcant (p ≤ 0.01) for all the parameters studied in both conditions (Table 1).2).To validate the above-mentioned fndings in diferent mutants, correlation between agronomic/yield-related parameters was determined (Table 3).In the control condition, productive tillers, total weight, and yield showed a negative association with plant height.Total spikelets and empty spikelets showed negative, whereas total weight, fertility, and yield showed positive association with productive tillers, and fertility and yield showed highly positive association with total weight in both sets.Te analysis of variance (ANOVA) for mean squares indicated that highly signifcant diferences were observed among all mutants for all the characters studied in both conditions as shown in Table 4.

Physio-Biochemical
Profling.A bit diferent trend was observed for physio-biochemical activities among all mutants under both control and stress conditions.A signifcant variation was seen in enzymatic antioxidants and chlorophyll content in both control and stress conditions.A signifcant decrease was also observed in all physio-biochemical assays among all mutants.Under control conditions, the ascorbate peroxidase (APX) content was observed higher (106.07 Units/g.f. wt.) in mutant NMSF-49 followed by NMSF-40 (105 Units/g.f. wt.).NMSF-6 depicted the lower content for APX (13.13 Units/g.f. wt.).NMSF-29 showed higher value (17.54 Units/g.f. wt.) and NMSF-20 showed lower value (4.64 Units/g.f. wt.) for catalase (CAT) activity.

Cluster Analysis.
For all mutants, a tree diagram was constructed through the hierarchical cluster analysis by using all the germination, seedling, yield, quality, and biochemical data under both control and water stress conditions.Cluster analysis categorized seventy-one mutants along with their parent line RICF-160 and commercial variety Kainat into fve clusters as shown in Table 6 and Figure 1.Cluster I contained maximum (37) number of mutants.Ten mutants were grouped in cluster II.Twentytwo were grouped in cluster III.Mutant NMSF-59 was placed into cluster IV, while three mutants NMSF-60, NMSF-62, and NMSF-64 were grouped in cluster V.

Correlation Analysis. Correlation (Pearson
) for all growth, seedling, yield, and physio-biochemical traits under both control and drought stress was carried [48] out using XLSTAT 2023, with 95% confdence interval (Figure 2 and Supplementary Table 2).Values in the bold form are different from 0 with a signifcance level alpha � 0.05.Values with negative sign showed a negative correlation.

Scientifca
Under the control condition, germination percentage (GP%) at 48 hours was positively correlated with germination percentage at 72, 96, and 120 hours, germination rate (GR), coefcient velocity of germination (CVG), seed vigor (SV), dry weight (DW), and empty spikelets (ES) in both conditions, while with peroxidase (POD) and total spikelets (TS) only in the control condition and in the stress condition, it was positively correlated with seedling height (SH), shoot length (SL), root length (RL), and fresh weight (FW).Germination percentage at 48 hours negatively correlated with the fertility percentage (F %) in control condition.Germination rate in both control and stress conditions was positively correlated with all the parameters except seedling height (SH), shoot length (SL), root length (RL), fresh weight (FW), ascorbate peroxidase (APX), and empty spikelets (ES) and also no correlation with chlorophyll (Chl) content, productive tillers (PT), panicle length (PL), total spikelets (TS), total weight (TW), and yield (Y) in both control and stress conditions.It was negatively with plant height (PH) and fertility (F %) in the control condition.
Shoot length under the control condition showed a positive correlation with seed vigor (SV), seedling height, root length, fresh weight, and dry weight in both conditions, while with germination percentage at 72 hours only in control and germination percentages at 96 and 120 hours and coefcient velocity of germination (CVG) only in stress condition.It was negatively correlated with productive tillers (PT) and total weight in control condition and showed no correlation with other parameters.Tere was no correlation between shoot length in the stress condition and fresh 8 Scientifca weight, chlorophyll content, ascorbate peroxidase activity (APX), catalase activity (CAT), peroxidase (POD), plant height, productive tillers, panicle length, total spikelets, empty spikelets, total weight, fertility, and yield.Results showed that there was no association between root length in both conditions and panicle length, total spikelets, empty spikelets, total weight, fertility, and yield while negatively correlated with productive tillers.Plant height under the control condition showed no correlation with all parameters except panicle length while negatively correlated with germination percentage at 48, 72, and 96 hours, germination rate, coefcient velocity of germination, dry weight and the stress tolerance index of the plant height, panicle length, and empty spikelets.On the other hand, under the stress condition, plant height positively correlates with panicle length, total spikelets, fresh weight, ascorbate peroxidase, and peroxidase in both conditions.It was noted that productive tillers under the stress condition showed positive correlation with total weight, fertility, and yield in both conditions, while negatively correlated with root length, seedling height, seed vigor, and ascorbate peroxidase activity in control and germination percentage at 48, 72, 96 hours, germination rate, and coefcient velocity of germination in stress condition.Empty spikelets, peroxidase activity, catalase activity, chlorophyll content (Chl), and fresh weight (FW) in both conditions showed a negative correlation with productive tillers under stress conditions.
Results revealed yield under control conditions showed a positive correlation with germination percentage at 72 and 96 hours, coefcient velocity of germination in control condition, stress tolerance index (STI) of ascorbate peroxidase, productive tillers, panicle length, fertility, and total weight in both conditions, while negatively correlated with chlorophyll content in stress, empty spikelets in control, and STI of total weight.Among physio-biochemical parameters, chlorophyll content positively correlates with germination percentage at 48 hours, germination rate, total spikelets, empty spikelets, POD activity, CAT activity, root length, fresh weight, and dry weight, under both conditions.Ascorbate peroxidase activity under the control condition showed a positive association with germination percentage at 48 hours, germination rate, coefcient velocity of germination, seed vigor, root length, fresh weight, plant height, total spikelets in stress conditions, and in both conditions with dry weight, catalase activity, and peroxidase activity, while negatively correlates with productive tillers and total weight.

Principal Component Biplot Analysis.
For a better understanding of the relationship among mutants and to extract the important and useful information present in the data matrix, the principal component analysis was performed [48] for all mutants and 24 traits with their stress tolerance index under both control and stress conditions.It also minimized the number of traits that describes the maximum percentage of variability present in the data.Te eigenvalue is very crucial that decides which principal components are important and useful for further study.Te highest eigenvalue refected as the best representative of system with the aspects of principal components [49].Under both conditions, 17 principal components (PCs) depicted more than 1.0 eigenvalue and explained 88.343% variability.Te frst fve components were most persuasive: PC-I contributed 25.861% of total variability; PC-II, PC-III, PC-IV, and PC-V individually contributed 10.745%, 6.955%, 6.468%, and 6.028%, respectively, while the cumulatively PC-I, PC-II, PC-III, PC-IV, and PC-V contributed 25.861%, 36.606%,43.561%, 50.029%, and 56.057% respectively, of the total variability (Supplementary Table 1).
Out of 24 traits under both conditions, 22 parameters showed positive factor loading in PC-I, while germination percentage (at 48, 72, 96, and 120 hours), germination rate, coefcient velocity of germination, seed vigor, and seedling height of both conditions (control and stress) and stress tolerance index (STI) had the greatest efect in PC-I.In PC-II, 16 parameters exhibited the positive factor loading with productive tillers, total weight, shoot length, root length, fertility, and yield having the greatest efect (Supplementary Table 1).For more authentic identifcation of mutants with maximum values for one or more traits, mutant-by-trait biplot was constructed against PC-I and PC-II for all the mutants and all the traits under both conditions (Figure 3).It explained the trait description of a mutant [50].Vector line was drawn from the origin of the biplot to understand the interrelation between mutants and traits.Genotypic performance that difers one mutant from the others can be assessed by the distance of the mutant from the origin of the biplot.Te more distant mutants could have more values for one or more traits.A scree plot showed cumulative variability and eigenvalues for studied parameters (Figure 4).

Discussion
In the present scenario of climate change, drought is one of the most severe abiotic stresses hampering seed germination, plant growth, and crop production [51][52][53].A high rate and consistency of germination under water stress conditions is very much important for good crop growth to achieve good productivity [54].Among diferent methods of breeding, mutation breeding has gained popularity among researchers due to its utilization in plant biotechnology because of certain limitations of diferent methods of breeding like hybridization and transgenic techniques [55].Among diferent mutagens, physical mutagens (gamma rays) are widely used due to their easy handling as compared to chemical mutagens, which may cause carcinogenic efects [56].Due to the irradiation of seeds, reactive oxygen species (ROS) or free radicals are generated in cells and cause random mutations [57].To create genetic variability in the desired variety for the trait of interest, the seeds of the parent varieties RICF-160 and RICF-159 and RICF-152 were exposed to diferent doses of gamma rays [58].In this study, we examined and matched the diference among mutant lines in control and water stress conditions [59].
In the present study, the frst experiment was conducted on 71 rice mutants along with their parent line RICF-160 and commercial variety Kainat was subjected to PEG-induced drought stress for the estimation of germination and seedling growth.Eight mutants were created from 200 Gy dose, eleven from 250 Gy dose, and fve from 300 Gy dose of RICF-160.Two mutants were created from 200 Gy dose, one form 250 Gy dose and two from 300 Gy dose of RICF-159.Forty-two mutants were created from 300 Gy dose of RICF-152 (Table 7).During the last few decades, mutagenesis has resurfaced in plant breeding.Plant mutagenesis, which produces new variety in crop plants, combined with in vitro selection and plant biotechnology methods enables breeders to select for characters that were previously difcult to achieve in breeding program.
Oxidative stress is regarded as a major damaging factor in plants [76] and animals [9,77] exposed to a variety of abiotic stresses including drought [78].So, it is important to examine the relationship between the imposition of drought stress and the induction of oxidative stress in the partial halophytic crop rice [79][80][81][82].APX appears to constitute a basic mechanism of deployment for antioxidative defense in plants [47,83].Our results indicated a decline in CAT activity under drought stress, which suggests, at least here, that CAT appears not to be an efective scavenger of H 2 O 2 in 10 Scientifca our case [84].Such observations suggest that increased water defcit induced severe oxidative stress in rice plants, where antioxidant defense system seemingly fails to combat the oxidative damage [85].   1 flter paper as shown in Figure 5. Sterilized PEG solution and distilled water were used as growth medium for stressed and control sets, respectively.Seeds in each Petri plates were placed at sufcient distance to allow the optimum growth of shoot and root length.Each plate was irrigated with 1 mL of autoclaved water (in case of control) and 1 mL of 16% PEG-6000 (in case of stress) with a pipette on a daily basis to moisten the flter paper, and the plates were placed in the dark to allow germination at 37 °C [86,87].Te experiment was continued for 12 days from the day of seed sowing for the control set and 15 days for the stress set as the water stressinduced late germination of seeds.After 15 days, the experiment was harvested for data recording.

Germination Attributes.
Te data for germination and germination-related parameters were collected on a daily basis.Te data were recorded for diferent germination parameters, namely, germination percentage (GP) and coefcient of velocity of germination (CVG), which give the indication of the rapidity of germination, and germination rate (GR) basically gives an idea of the percentage of seeds germinating per day.
4.1.4.Seedling Attributes.Tree seedlings were selected randomly from each replicate for the data collection of growth parameters such as root length (RL), shoot length (SL), seedling height (SH), seed vigor (SV), fresh weight (FW), and dry weight (DW) following the earlier worker [43].Te seedlings with 2 mm of RL were measured as germinated [88].Te root length, shoot length, and seedling height were measured in cm by using a scale.Fresh weight was measured using an analytical balance.Te seedlings were then oven-dried at 70 °C for 24 h for the estimation of the dry weight.
Table 7: Details of the experimental material along with their parent line and commercial variety (CV) and gamma radiation doses from which these mutants were originated.Two water regiments were used and irrigated and water stress conditions using randomized complete block design with three replicates for both conditions.Forty-day-old seedlings were transplanted 20 cm apart between rows and 15 cm within the rows.Irrigation was stopped for one of the sets at the initiation of booting stage to create the drought stress and last till maturity.All necessary precautions were taken to maintain uniform plant population in each treatment per replication.

Plant Culture and Harvest.
All the recommended practices were followed along with the plant protection measure to raise a good crop.Observations were recorded, and the data were subjected to statistical analysis.Ten, the samples were collected at fag leaf stage in labeled zipper bags from both normal and stressed sets for antioxidant estimation.Te samples were then stored at −20 °C to ensure the preserve integrity.Te analysis was carried out at Marker Assisted Breeding (MAB) Lab-1, Plant Breeding and Genetics Division, Nuclear Institute for Agriculture and Biology (NIAB), Faisalabad, Pakistan.

Physiological and Biochemical Parameters
(1) Chlorophyll Content.Chlorophyll content was measured as SPAD (soil plant analysis development) value from the third upper leaf of the three plants [89].
(2) Antioxidant Enzymes.Fresh leaves (1 g) were ground in 1.5 ml (50 mM) potassium phosphate bufer (pH 7.4).Samples were then centrifuged at 14000 rpm for 10 minutes at 4 °C.Te supernatant was separated and used for the determination of the diferent enzymatic and nonenzymatic activities.All the data were taken in triplicate.Te extract was used for the assay of the following antioxidant activities as described earlier [48].
(3) Ascorbate Peroxidase (APX) Activity.APX activity was measured by using 50 μl sample extract, 1000 μl H 2 O 2 , and assay bufer (10 mM ascorbic acid 3.4 ml, 500 mM EDTA 10 ml, 200 mM potassium phosphate bufer 25 ml, and distilled water 50 ml) 1000 μl [90].Te reaction was initiated by the addition of 1 ml of 10% (v/v) H 2 O 2 , and the oxidation rate of ascorbic acid was estimated by the following of the decrease in absorbance at 290 nm for 1 min [91].
(4) Catalase (CAT) Activity.CAT was estimated by the following method described by Beers and Sizer [92].For the measurement of CAT activity, assay solution contained 50 mM phosphate bufer (pH 7.4) 2 ml, 59 mM H 2 O 2 , and 100 μl and enzyme extract 100 μl.Te decrease in the absorbance of the reaction solution at 240 nm was recorded after every 20s for 1 minute.
(5) Peroxidase (POD) Activity.Activity of POD was measured using the method of Chance and Maehly [93] with some modifcation.For the measurement of POD activity, (6) Statistical Analysis.Te screening experiments were conducted in three repeats using a randomized completely block design (RCBD) [48].Te signifcance was determined by ANOVA and Tukey (HSD) test at p < 0.05 by using XLSTAT 2023 software.To check the response of the mutants under control and stress (drought) treatments, bar graphs were constructed based on mean ± S.E.In graphs, bars with diferent alphabets were signifcantly diferent from each other.Principle component analysis (PCA) for all the parameters under both conditions was performed, and for the frst two principal components (PC-I and PC-II), biplots were constructed by using the same software.Correlation (Pearson test) and cluster analysis were also performed by algometric hierarchical clustering for all mutants under all traits by using the same software.

Figure 2 :
Figure 2: Correlation matrix showing Pearson's correlation among germination, seedling, yield, and physio-biochemical traits for diferent rice mutants under control and drought stress conditions.

10 Figure 3 :
Figure 3: Biplots of frst two principal components (PC-I and PC-II) for control and stress conditions.

Figure 4 :
Figure 4: Scree plot representing cumulative variability and eigenvalues for studied parameters.

4. 2 .
Field Trial 4.2.1.Experimental Design and Treatments.Te same seventy-one mutants along with parent line RICF-160 and commercial variety Kainat were used in the feld experiment.Te experiment was conducted in the experimental feld area at the Nuclear Institute for Agriculture and Biology, Faisalabad.

Figure 5 :
Figure 5: Seed germination after 15 days in both control "C" and drought "D" conditions.

Table 1 :
Mean squares from analysis of variance under control and stressed condition.

Table 2 :
Descriptive statistics of yield attributes of rice mutants.

Table 3 :
Pair-wise correlation between yield attributes of mutants.

Table 4 :
Mean square of yield attributes of both (control and stress) sets.

Table 5 :
ANOVA and mean square for physio-biochemical traits of both (control and stress) sets.