First Report of Sesame Mutants Tolerant to Severe Drought Stress during Germination and Early Seedling Growth Stages

In the context of climate change and water scarcity, there is a need to develop and use drought-tolerant sesame cultivars. This study was conducted to evaluate the response of 13 sesame genotypes, including 11 mutants and their wild-types, to drought during germination and early seedling growth. Moderate and severe drought stress was simulated by applying polyethylene glycol (PEG) at two osmotic potentials, −0.6 MPa and −1.2 MPa, respectively, on seeds of two successive mutant generations, M2 and M3. The parameters measured or calculated were germination percentage (GP), germination rate (GR), mean germination time (MGT), root length (RL), shoot length (SL), root to shoot ratio (RSR), and the seedling vigor index (SVI). Results showed the significant effect of genotype, drought, and drought × genotype interaction on all parameters investigated. Under severe drought, seeds of seven genotypes, including wild types, were not able to germinate. There was a drastic decline of all parameters for the rest, except MGT and RSR, which markedly increased. Interestingly, two mutants, “ML2-5” and “ML2-10”, were identified as the most tolerant to severe drought and the most stable over both generations. The present work is the first report of sesame germplasm with such a high level of tolerance to drought during germination and early seedling growth stages.


Introduction
Sesame (Sesamum indicum L.) is known as "the queen of oilseeds" due to the high oil content and nutritional quality that characterize its seed [1]. Sesame seeds are rich in oil, protein, carbohydrates, vitamins, nutrients, antioxidants, and minerals as important nutritional sources for human health [2,3]. Sesame is a tropical and subtropical crop, but it is also cultivated under arid and semiarid climate conditions [4]. In 2019, the world's cultivation area of sesame was around 12.82 Mha, ensuring a world production of about 6.55 Mt, of which about 60% comes from Asia [5].
Unlike other oilseed crops, sesame is reported to be more tolerant to drought [6]. However, in arid and semiarid areas, drought often occurs conjointly with heat or high temperatures and impairs sesame production significantly [2,3]. Harmful effects on sesame seed yield and quality are markedly noticed when this water stress happens, especially at germination and flowering stages [7][8][9][10]. Severe or prolonged drought adversely influences sesame productivity by reducing the number of capsules per plant, the yield, and the quality of the oil [11][12][13]. Drought stress can also affect the level of secondary metabolites and morphophysiological characteristics of the sesame seed [14]. Besides, drought becomes more damaging during flowering as it increases the susceptibility of the plant to attack from pathogens [15]. Seed germination is the first critical and most sensitive stage of the plant life cycle [6,16,17], because of its direct and strong correlation with the seedling

Drought Stress Effects on Early Seedling
ANOVA results showed, in both generations, a significant effect (p < 0.05) of drought, genotype, and drought × genotype interaction on all early seedling growth parameters, namely shoot length (SL), root length (RL), root-to-shoot ratio (RSR), and the seedling vigor index (SVI) ( Table 1). However, among both generations, significant variation was only recorded for RSR and the SVI, suggesting there was a stability of the genotypes studied for SL and RL over M2 and M3 generations. Figure 4 presents the mean values of shoot length (SL) in the presence and absence of water deficit. In the absence of stress (control), the highest mean SLs of 5.90, 5.50, and 5.48 cm were observed in "US1-DL", "US1-2", and "US2-1", respectively. In the presence of drought, SL decreased in all genotypes. At a moderate water potential (−0.6 MPa), the most remarkable shoot length (about 2 cm) was observed in "ML2-10" and "US1-3", while the shortest SL (about 1 cm) was found in "US1-DL", "US2-6", and "US2-7" in both generations ( Figure 4A). In the case of severe water stress (−1.

Drought Stress Effects on Early Seedling
ANOVA results showed, in both generations, a significant effect (p < 0.05) of drought, genotype, and drought × genotype interaction on all early seedling growth parameters, namely shoot length (SL), root length (RL), root-to-shoot ratio (RSR), and the seedling vigor index (SVI) ( Table 1). However, among both generations, significant variation was only recorded for RSR and the SVI, suggesting there was a stability of the genotypes studied for SL and RL over M 2 and M 3 generations. Figure 4 presents the mean values of shoot length (SL) in the presence and absence of water deficit. In the absence of stress (control), the highest mean SLs of 5.90, 5.50, and 5.48 cm were observed in "US1-DL", "US1-2", and "US2-1", respectively. In the presence of drought, SL decreased in all genotypes. At a moderate water potential (−0.6 MPa), the most remarkable shoot length (about 2 cm) was observed in "ML2-10" and "US1-3", while the shortest SL (about 1 cm) was found in "US1-DL", "US2-6", and "US2-7" in both generations ( Figure 4A). In the case of severe water stress (−1.2 MPa), the two mutants "ML2-10" and "ML2-5" confirmed their early growth potential and stability under stressful conditions. Shoots of "ML2-10" reached 1.52 and 1.16 cm length in generations M 2 and M 3 , respectively, while those of "ML2-5" outreached 1.06 and 1.23 cm, respectively ( Figure 4B). "ML2-10" and "ML2-5" confirmed their early growth potential and stability under stressful conditions. Shoots of "ML2-10" reached 1.52 and 1.16 cm length in generations M2 and M3, respectively, while those of "ML2-5" outreached 1.06 and 1.23 cm, respectively ( Figure 4B). Regarding root length (RL), and in the absence of any water stress, the genotypes "US1-DL" and "ML2-5" exhibited the highest average values, 4.70 and 4.52 cm, respectively. In contrast, "US06" and "US2-6" recorded the lowest average RLs, 3.32 and 3.35 cm, respectively ( Figure 5). Overall, RL increased under moderate drought conditions in all genotypes and for both seed generations, particularly in "ML2-10" and "ML2-5", which maintained their superiority (more than 6 cm) in both M2 and M3 generations ( Figure 5A). For severe water stress, a strong decrease in RL was noted in all genotypes ( Figure 5B). Again, and unlike the other genotypes, "ML2-5" and "ML2-10" showed and retained the lowest reduction in RL of 47 and 46%, respectively, in the M2 generation, and 53 and 52%, respectively, in the M3 generation. This indicates that "ML2- Regarding root length (RL), and in the absence of any water stress, the genotypes "US1-DL" and "ML2-5" exhibited the highest average values, 4.70 and 4.52 cm, respectively. In contrast, "US06" and "US2-6" recorded the lowest average RLs, 3.32 and 3.35 cm, respectively ( Figure 5). Overall, RL increased under moderate drought conditions in all genotypes and for both seed generations, particularly in "ML2-10" and "ML2-5", which maintained their superiority (more than 6 cm) in both M 2 and M 3 generations ( Figure 5A). For severe water stress, a strong decrease in RL was noted in all genotypes ( Figure 5B). Again, and unlike the other genotypes, "ML2-5" and "ML2-10" showed and retained the lowest reduction in RL of 47 and 46%, respectively, in the M 2 generation, and 53 and 52%, respectively, in the M 3 generation. This indicates that "ML2-10" and "ML2-5" are efficient and stable in developing high root system growth even in the presence of severe drought.
Finally, average values of all studied parameters, regarding both seed germination and early seedling growth, are shown in the supplementary material Table S1.

Drought Stress Effects on Germination
In the present study, all measured seed germination parameters, GP, GR, and MGT, were affected by drought in all genotypes studied, with a decrease in GP and GR and an increase in MGT, as a response to stress exposure. This is in agreement with findings of previous studies in sesame [8,12,19,20], as well as in other oilseed crops like rapeseed [28,29], sunflower [30], and safflower [21]. This may be due to the alteration of enzymes and hormones present in the seed or the generation of free radicals, which alter the metabolic pathways in seeds germinating under drought stress [31,32]. As a result of pronounced seed dehydration, there is an alteration of mechanisms leading to embryo development [33]. In fact, seed germination is directly linked to the use of reserves, res-

Drought Stress Effects on Germination
In the present study, all measured seed germination parameters, GP, GR, and MGT, were affected by drought in all genotypes studied, with a decrease in GP and GR and an increase in MGT, as a response to stress exposure. This is in agreement with findings of previous studies in sesame [8,12,19,20], as well as in other oilseed crops like rapeseed [28,29], sunflower [30], and safflower [21]. This may be due to the alteration of enzymes and hormones present in the seed or the generation of free radicals, which alter the metabolic pathways in seeds germinating under drought stress [31,32]. As a result of pronounced seed dehydration, there is an alteration of mechanisms leading to embryo development [33]. In fact, seed germination is directly linked to the use of reserves, respiration, and phytohormones that are all affected during the development of the embryo in the stressed seed [34,35]. Therefore, the development of genotypes with higher seed metabolic efficiency under drought conditions is a desirable crop improvement trait. In our case, all genotypes were slightly affected by moderate water stress. However, under severe drought conditions, their seed germination was drastically reduced. In some of these genotypes, germination was inhibited in both generations, M 2 and M 3 . These results are in agreement with those of Boureima et al. [8], Harfi et al. [20], and Dissanayake et al. [12], who reported that the germination of sesame seeds is completely inhibited by a water potential lower than -1 MPa of PEG. However, some mutants kept germinating in this particular situation, showing their tolerance to such a high moisture stress level. Among them, the mutants "ML2-5" and "ML2-10" are the most interesting as they were stable over both generations M 2 and M 3 , and the least affected, with a respective average reduction of less than 63 and 70%. This is, so far, the first report of sesame genetic materials that are able to germinate at an osmotic potential of −1.2 MPa. Germination rate (GR) is one of the indices used to evaluate crop tolerance to drought stress. A reduction in germination rate and total germination is one of the most common responses of drought-exposed plants [36,37]. Our results showed that GR decreases significantly with water deficit, which is in agreement with the results of El Harfi et al. [20] and Bakhshandeh et al. [38] in sesame. In the present study, the mutants "ML2-5" and "ML2-10" had the highest GR values, over both generations, M 2 and M 3 , under moderate (80% and 75%, respectively) and severe (40% and 25%, respectively) moisture stress.. The mutant "ML2-5" is of particular interest as it exhibited not only the highest germination rate but also the lowest mean germination time (MGT) under severe drought conditions. It is well known that in such conditions, the seed requires more time to adjust its internal osmotic potential to the external environment [39] and, thus, takes more time to germinate compared to non-stressful conditions. However, a delay in the MGT can be harmful to the successful establishment of a crop stand. Therefore, and based on all seed germination parameters studied, the mutant "ML2-5", followed by the mutant "ML2-10", is the most tolerant to high levels of moisture stress. They both could be considered as novel and relevant drought-tolerant germplasms, during the germination stage, since no similar sesame material has been found and reported in the literature.

Drought Stress Effects on Early Seedling
A remarkable diminution in shoot length was observed with a decrease in water potential from 0 to −1.2 MPa. This may be due to the high sensitivity of shoot tissues to water deficit [40]. Similar results were reported in sesame by Mensah et al. [19], El Harfi et al. [20], and Dissanayake et al. [12]. Our findings show that "ML2-10" and "ML2-5" are the most drought-tolerant among the genotypes studied. In fact, under moderate stress, "ML2-10" and "ML2-5" developed SLs of about 2 cm in both M 2 and M 3 generations, and even under severe stress these two mutant lines maintained SLs above 1 cm over the two generations. In previous studies on sesame, no shoot growth was reported for similar severe drought conditions. Additionally, root length was drastically reduced by severe water stress (−1.2 MPa). However, it was improved under moderate stress, compared to non-stress conditions (control). The decrease in root length could be due to reduced cell multiplication in root meristems [41]. Similar findings were reported by Boureima et al. [8], who described that root length increased under moderate stress (−0.5 MPa) and decreased at severe stress (−1 MPa) in Senegalese sesame. Likewise, other previous studies have shown a significant reduction in root length under severe stress [12,19,20,42]. In our research, the longest roots were found in "ML2-10" and "ML2-5", both under severe and moderate stress and over both generations. These findings are interesting in terms of early drought-tolerance and genetic stability. Admittedly, root traits are the first to be affected under drought stress conditions, and genotypes exhibiting better performance may be more tolerant [43]. Additionally, a plant's ability to develop an extensive root system contributes to its drought tolerance [44]. Thus, root morphology and/or growth rate may be promising markers for selecting drought-tolerant varieties [45,46].
The root-to-shoot ratio (RSR) reflects the manner in which roots develop with regard to the growth of a plant. A high RSR indicated the faster growth of the root compared to the shoot. Our results showed that moderate and severe stresses led to the increase of this ratio in most of genotypes. The RSR for the genotypes tested varied considerably, "ML2-5", "US1-3", and "ML2-10" were the most tolerant to moderate and severe drought through both M 2 and M 3 generations, as they exhibited the highest RSR values. Finally, the seedling vigor index (SVI), combining germination and shoot growth, is more sensitive to drought and could be considered an effective indicator of drought tolerance in crops [47]. Our findings showed a diminution in the SVI after seedling's exposure to drought, which is in agreement with reports of Spielmeyer et al. [48] in wheat and Koskosidis et al. [49] in chickpeas. The mutant lines "ML2-5" and "ML2-10" are the least affected, maintaining the highest SVI under both moderate and severe stresses and confirming, thus, their highest tolerance to moisture stress.
Finally, significant differences were recorded between the two generations (M 2 and M 3 ) for GP, RSR, and the SVI, particularly in the mutants "US2-7", "US1-2", and "US1-3". This could be explained by genetic factors related to these genotypes and pointed to the instability of the mutated genes related to these parameters. Contrarily, in most of the mutants studied, one could observe slight differences among both generations for the majority of the parameters investigated. This is particularly true for the most drought-tolerant mutants, "ML2-10" and "ML2-5", which suggests their genetic stability and confirms their suitability to be used as relevant and appropriate germplasm in breeding programs aimed at improving drought tolerance in sesame during germination and early seedling growth stages.

Plant Material
This study's plant material consists of 11 sesame mutant lines, along with their two wild-types ("ML13" and "US06"). Using chemical EMS-mutagenesis, these mutants were recently developed and characterized as described by Kouighat et al. [25]. They were selected based on some phenological, morphological, and agronomic traits. Their most important characteristics are summarized in Table 2. In the present work, seeds from both M 2 and M 3 generations were studied to confirm the results obtained and to assess the stability of the characters and behaviors observed in such mutant lines.

Seed Treatment under PEG-6000 Solution
The experiment was carried out in 2020 at the oilseed crops laboratory in the Regional Center of Agricultural Research in Meknes, belonging to the "Institut National de la Recherche Agronomique" (INRA-Morocco). It was conducted in a completely randomized design, with two factors and three replications. The first factor was the genotype (mutants and parents), with 13 levels, and the second one was the drought stress, induced by the PEG solutions, with three levels of water potential, 0, −0.6, and −1.2 MPa, applied according to the equation of Michel & Kaufmann [50]. These potential levels were designed to simulate absence of water stress, moderate stress, and severe stresses, respectively. The choice of water potentials was based on previous studies showing that the use of the osmotic potential −1.2 MPa completely inhibited the germination of sesame seeds [8,20,21]. The seeds used in this study, from both M 2 and M 3 generations, were harvested in 2019 and 2020, respectively. For each genotype, 50 sesame seeds were immersed in 5% sodium hypochlorite for 5 min and then rinsed with distilled water to sterilize their surfaces. The seeds were incubated for germination on Whatman paper in glass Petri dishes (10 cm × 15 mm) at 24 ± 1 • C in the incubator. Every 48 h, 3 mL of PEG-6000 solution were added to the Petri dishes. However, in control (0 MPa), 3 mL of distilled water were added instead. The seeds were considered to be germinated when the radicle reached up 2 mm length [51].

Parameters Calculated/Measured and Statistical Analyses
The germinated seeds were counted at regular intervals every 48 h until the end of the experiment (eight days). The germination percentage (GP), calculated on day eight, as well as the germination rate (GR), were determined as follows: GP = (N8/50) × 100, and GR = Σ [(Gi − Gi − 1)/i], where N8 is the number of seeds germinated on the 8th day, Gi is the number of seed germinated on the day i, and Gi-1 is the number of seeds germinated on day i − 1 [52].
To determine statistically significant differences between genotypes, drought, and their levels of interaction, data were subjected to analysis of variance (ANOVA) with two factors, all considered as fix. Duncan's new multiple range test (DMRT) was applied to compare treatment options and classify genotypes according to their tolerance to drought. These analyses were performed using the software package SPSS for Windows (version 20).

Conclusions
In conclusion, this is the first report of sesame mutants which are highly droughttolerant and stable during seed germination and early seedling growth, managing to germinate and grow under severe drought corresponding to a water osmotic potential of −1.2 MPa. Nevertheless, an experiment under field conditions, with an irrigation deficit regime, would be useful to confirm such findings. The two mutants, "ML2-5" and "ML2-10", could be managed and used as promising germplasms for developing cultivars with a high potential of germination under limited water availability conditions. Additionally, it would be interesting to assess the tolerance of these genotypes to drought stress occurring at different adult plant stages, particularly during flowering and grain-filling periods.