https://orcid.org/0000-0002-3980-9822
Figures
Abstract
Climate change is projected to increase the incidence of severe drought in many regions, potentially requiring selection for different traits in crop species to maintain productivity under water stress. In this study, we identified a suite of hydraulic traits associated with high productivity under water stress in four genotypes of S. melongena L. We also assessed the potential for recovery of this suite of traits from drought stress after re-watering. We observed that two genotypes, PHL 4841 and PHL 2778, quickly grew into large plants with smaller, thicker leaves and increasingly poor hydraulic status (a water-spender strategy), whereas PHL 2789 and Mara maintained safer water status and larger leaves but sacrificed large gains in biomass (a water-saver strategy). The best performing genotype under water stress, PHL 2778, additionally showed a significant increase in root biomass allocation relative to other genotypes. Biomass traits of all genotypes were negatively impacted by water deficit and remained impaired after a week of recovery; however, physiological traits such as electron transport capacity of photosystem II, and proportional allocation to root biomass and fine root length, and leaf area recovered after one week, indicating a strong capacity for eggplant to rebound from short-term deficits via recovery of physiological activity and allocation to resource acquiring tissues. These traits should be considered in selection and breeding of eggplant hybrids for future agricultural outlooks.
Citation: Delfin EF, Drobnitch ST, Comas LH (2021) Plant strategies for maximizing growth during water stress and subsequent recovery in Solanum melongena L. (eggplant). PLoS ONE 16(9): e0256342. https://doi.org/10.1371/journal.pone.0256342
Editor: Federico Brilli, National Research Council of Italy (CNR), ITALY
Received: January 7, 2021; Accepted: August 5, 2021; Published: September 1, 2021
This is an open access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication.
Data Availability: All raw trait value data files are available from the Ag Commons Database (https://data.nal.usda.gov/) at DOI: 10.15482/USDA.ADC/1520948.
Funding: E.D. was supported by a Fulbright-Philippine Agriculture Advance Research Grant sponsored by the Philippine Department of Agriculture, the US Department of State, Bureau of Educational and Cultural Affairs and the Philippine American Educational Foundation to conduct this research in Colorado. L.H.C. was supported under project NP211 3012-13000-010-00D at the USDA Agricultural Research Service.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Water shortages are responsible for the greatest crop losses around the world and expected to worsen in many regions where agricultural water availability is challenged with precipitation declines due to climatic changes, and/or competition with municipal and industrial needs [1, 2]. To improve crop production under water limitations we need a foundational understanding of traits contributing to crop performance and crop strategies that maintain productivity [3–6]. Furthermore, although traits that drive crop yields alter performance late in plant development, they may be correlated with trait responses to water limitations at earlier growth stages [7].
Many plant processes respond concurrently when plants are under water limitations. However it has been clearly observed that even within the same species, different traits may support or damage crop productivity under water stress [8, 9]. Additionally, the timing and duration of water stress can activate differing trait responses [8, 9]. Despite this, strategies are considered to fall into two broad categories: traits that conserve water (water savers) versus traits that exacerbate declining hydraulic function (water spenders) [10]. In wild ecosystems, both strategies can result in increased survival and reproductive success; “drought tolerance” involves the expression of water saver traits to withstand water deficits until rain events, whereas “drought avoidance” is typical of water spenders that grow and reproduce rapidly before the onset of drought [11–14]. Given that survival alone is a poor benchmark for crops, desirable crop genotypes for drought-prone systems must also maintain high productivity under water limitations [3, 5, 15, 16]. While water savers may effectively limit water loss by closing stomata [17, 18] and engaging in gas exchange only briefly in the morning and at night, this may result in the undesirable reduction of photosynthesis, growth rates, and yield [19–22]. Furthermore, species that cope with low tissue water potential by maintaining turgor through osmotic adjustment, increased cell elasticity, reduced cell size, and protoplasmic resistance often expend energy that reduces productivity [23, 24]. In some species, fast growing water spenders manage to maximize productivity without protecting tissue turgor or hydraulic safety (as do “early vigor” varieties in many species) [25, 26]. In these varieties, the same traits supporting optimal growth under well-watered conditions may also support optimal growth under water deficit [27]. Water spenders with the potential to “avoid” drought and still produce a viable crop may do so by increasing root depth or developing a more efficient root system for water foraging [24, 28–30]. The growth of a deep and extensive root system can come at the expense of larger shoot systems and high yields [31, 32]. Either way, the importance of root traits to crop performance under water stress has been evident; greater root length density can produce higher yields under drought [33].
The ability to recover from damage sustained during drought events varies substantially between species and cultivars and is as important for maintaining total seasonal productivity as traits that minimize damage during drought stress [34]. Some drought tolerance strategies can actually slow post-watering recovery; for example, roots exposed to dry soil can become suberized and limit water uptake until new root growth occurs [35, 36]. Key post-drought recovery traits seem to be focused on rapid restoration of photosynthesis (membrane repair, photosystem protection and repair, pigment protection and repair), hydraulic conductance (aquaporin up-regulation, cavitation reversal, rapid osmotic adjustments), and fine root flushing [37–40].
In the Philippines and much of southeast Asia, climate change is projected to increase the incidence and severity of typhoons but also the intensity of El Niño events, raising the occurrence of severe drought in a historically wet tropical region [41–43]. Due to their high water content, drought disproportionally impacts vegetable crops (as opposed to grains and tree fruits), which form a large component of Philippine agriculture [44]. A major strategy for weathering drought events is selection and cultivation of drought resistant genotypes and germplasm. International institutions such as the International Maize and Wheat Improvement Center (CIMMYT), the World Vegetable Center, the International Crops Institute for Semi-Arid Tropics (ICRISAT) and International Institute of Tropical Agriculture (IITA) have undertaken collections of diverse crop germplasm [45–48]. These groups have successfully screened many crop species, such as wheat and maize, for drought resistant genotypes [49]. Genotypic variability of drought tolerance has been established in nearly all crop species, from wheat, faba bean, and cotton [20, 50, 51] to watermelon [52]. The eggplant, Solanum melongena L., is ranked as one of the most important vegetables consumed and produced in the Philippines with an estimated production of 179,000 metric tons worth ~ US$32 million [53, 54]. S. melongena has been shown to be adversely affected by drought in terms of plant height, total dry weight and fruit yield [55–58]. Reduction in transpiration rate, stomatal conductance and photosynthetic rate of eggplant were also observed upon exposure to different drought stress durations [59]. Preliminary work with Philippine S. melongena genotypes showed a reduction of 21–29% in fruit production exposed to short-term drought stress [60].
In this study, we select four eggplant genotypes with a range of productivity under short duration water stress, including a reference variety [60]. Previous work has addressed drought- mediated shifts in photosynthetic pigments and reactive oxygen scavengers [57, 61] whereas we focus on traits that reflect the hydraulic functioning of S. melongena. We measured plant hydraulic, gas exchange, and morphological traits and identified correlations with biomass productivity under a two-week drought and subsequent one-week recovery period. We ask whether there is a tradeoff between water spender or saver strategies and vegetative biomass accumulation during the drought or recovery phase among the four genotypes.
Methods
Greenhouse study
A greenhouse experiment was set up to identify physiological traits associated with drought tolerance in eggplant. Solanum melongena genotypes PHL 4841, PHL 2778 and PHL 2789 were chosen based on drought performance in previous field and greenhouse trials [60] of 100 germplasm accessions from the National Plant Genetic Resources Laboratory in the Institute of Plant Breeding, University of the Philippines at Los Baños, Laguna, Philippines. S. melongena ‘Mara’, a released variety from the Institute of Plant Breeding, UP Los Baños was included as a reference variety. Seeds were sown into seedling trays containing fritted clay (Turface Greens Grade, Profile Products, Buffalo Grove, IL, USA) at the end of February in a greenhouse in Fort Collins, CO, USA. After 17–21 days, individuals were transplanted into 7.57 L plastic pots containing 10 kg fritted clay and watered to holding capacity via a drip irrigation before treatments were established. Pots were positioned on two greenhouse benches in a randomized complete block design of two factors: water availability (drought vs. well-watered control) and genotype (four genotypes). There were 5 replicates of each block (40 plants). This basic block design was doubled, and plants were harvested at 2 time points (post-drought and post-recovery) for a total of 80 plants.
Plants were maintained under a combination of natural sunlight and supplemental LED illumination on a 14:10 hour day:night cycle, corresponding to average temperatures of 22 and 29°C. Plants were fertigated using Grow More water soluble fertilizer (Grow More, Inc., Gardena, CA) amended with additional N in the form of urea and additional K in the form of KH2PO4 to achieve 79.5–22.5–5 ppm N:P:K daily for the first month after transplantation and transitioned to 60-30-120 ppm N:P:K for the remainder of the experiment.
Drought treatments began at 5 weeks after transplanting and lasted for 2 weeks for all blocks (80 plants). “Drought” plants received 30% of evapotranspired water (ET) of “control” plants daily for the first week, and 10% daily in the second week. “Control” plants were given 100% of ET daily. ET was calculated by weighing control pots daily at 14:00 hr to measure evaporative and transpiration water loss relative to 100% holding capacity. After the conclusion of the drought treatment, all remaining plants were re-watered to pot holding capacity.
Physiological measurements.
Sampling of drought and control plants were carried out from the 11th through the 15th and final day of the drought treatment on 50% of the experimental plant population. To avoid confounding genotype-specific traits with increased stress to experimental plants at the end of the sampling period, one replicate of each genotype was sampled each day. On each day, the third fully expanded leaf of each plant was measured for chlorophyll fluorescence (Fv/Fm) from 07:30 until 08:30 using a portable OS5P fluorometer (Opti-Sciences Inc., NH, USA). Each leaf was dark acclimated with leaf clips for 20 minutes prior to measurement. From 09:00 to 12:00 hrs, the same leaves were measured for photosynthetic rate (An), stomatal conductance (gs), and transpiration using the Li-COR 6400XT infrared gas analyzer with attached leaf measurement chamber (LiCOR Inc., Lincoln, Nebraska). Conditions in the leaf measurement chamber were the following: PAR (photosynthetically active radiation) of 1800 μmol m-2s-1, leaf temperature of 25°C, and CO2 concentration of 400 μmol mol-1. Instantaneous water use efficiency (WUEi) was calculated as the ratio of An to gs.
Leaf water potential (ΨL) was determined with use of a Scholander pressure chamber (Soil Moisture Equipment Corp., Santa Barbara, CA, USA). The same leaf used for gas exchange measurements was cut from each plant and immediately placed in a plastic bag in a cooler until ΨL could be measured (up to 1 hour).
After one week under full watering, “recovered” and control plants were again measured as above for chlorophyll fluorescence and leaf water potential.
Plant growth measurements.
Following physiological measurements, the aboveground portions of drought and control plants were cut and partitioned into leaves and stem. Total leaf area was measured for each plant using a Li-3100C leaf area meter (LiCOR Inc., USA). Partitioned shoot tissue was then oven dried at 60°C for 48 hours and weighed.
The belowground biomass of each plant was washed free of fritted clay and partitioned into fine and coarse roots. A representative sample of fine roots was obtained for each sample and stored in 30% ethyl alcohol for root scanning. Preserved fine roots were scanned in water in 2-D transparency mode with a desktop scanner (EpsonV750, Epson America Inc., USA) and analyzed using WinRHIZO™ software (Regent Instruments Inc., Canada). Remaining fine and coarse roots were dried and weighed as above. Leaf area ratio (total leaf area per total plant dry mass, m2 g-1; LAR) and specific leaf area (leaf area per leaf dry mass, m2 g-1; SLA), and leaf mass area (leaf dry mass per leaf area, g m-2; LMA) were calculated using the leaf data for each plant. Specific root length of fine roots (root length per dry mass, m g-1; SRLFineRts) and total root mass fraction (RMF, total root mass per total plant weight) were calculated using the fine root length and root biomass data from each plant [31, 62].
At the end of the recovery phase, all plants were also destructively sampled for measurement of leaf area and above- and below-ground biomass partitioning as above.
Field study
A field trial was conducted during the dry season of 2015 to evaluate the performance of 29 eggplant genotypes including Mara, PHL 4841, PHL 2778 and PHL 2789 at the experimental farm of the Institute of Plant Breeding, University of the Philippines Los Baños. The data and findings of the other 25 genotypes are presented in another manuscript. Plants were fertilized with a total of 170-70-180 kg N, P, K ha-1. Fertilizers were applied at planting, 3 weeks after transplanting (WAT), 6 WAT and during the fruiting stage. Three-week-old- seedlings of each genotype were transplanted in 4 row plots, 4 m in length, with 0.75 m distance between row and 0.5 m distance between plants.
The trial was laid out in a split plot design. Watering treatment (moisture stress and well-watered treatments as the main plot while the accessions/varieties were assigned to the smallest plots. The trial had 4 replications, with 4 row- plots measuring 4 meters in length, and 0.75 meter row distance.
Water stress was imposed by withholding irrigation among the water stress treatment plots at 5 WAT while the control plots were irrigated weekly through furrow irrigation. Water stress was imposed for two weeks and was terminated with rainfall occurrence after the 2nd week of irrigation witholding. After the heavy rain, stressed plants were allowed to recover with equal irrigation to control plots.
Six plant samples per genotype were collected for biomass partitioning, initial harvest, and root traits determination after termination of water stress treatment. Plant samples were partitioned into roots, stem, leaves and fruits. Leaves were cleaned and measured for leaf area using Li-3100C leaf area meter (Li-COR, Lincoln, Nebraska, USA). The roots were cleaned and washed free of soil particles prior to air drying for 2–3 days and eventually oven dried for 48 hours at 60°C together with above ground biomass. Dried samples were weighed on a top loading balance.
Two harvests were obtained during water stress phase while two harvests were collected at recovery. Mature fruits were harvested from 5 plants from inner and center rows and weighed immediately for total fresh weight.
Data analysis
Distributions of trait residuals were assessed for non-normality and transformed as needed. Traits for each ANOVA were examined for unequal variances within each treatment factor combination and the validity of differences between means of unequal variances were checked using Welch’s ANOVA and, when necessary, non-parametric Pair-wise Comparisons using Wilcoxon’s method. To assess the effect of water stress on S. melongena, all traits were analyzed with a Least-squares means test by Water stress treatment (control vs. water stress), with Genotype as a random effect (REML method). All analyses were performed in JMP v.13 except for VarClus cluster analysis (SAS) and PCAs (R).
To assess the differential response of the four S. melongena genotypes to water stress, all traits were analyzed with a one-way ANOVA with Genotype as the main effect. Finally, to determine and visualize which traits dominate the S. melongena water stress response and to understand trait partitioning among the four genotypes, we performed a Principal Component Analysis (PCA) in R using the prcomp() function in ggplot2. We chose the following nine variables for the PCA based on cluster analysis and a priori hypotheses of which traits might contribute to different drought-adaptive strategies: stem dry mass, fine root dry mass, An, transpiration rate, Fv/Fm, ΨL, SRL, SLA, and RMF.
To assess the effect of recovery post-water stress on S. melongena, all trait data measured after the recovery phase were analyzed with a Least-squares means test by Water stress treatment (control vs. water stress) with Genotype as a random effect (REML method). To assess the persistence of genotype-specific trait values for traits that displayed significant differences among genotypes during the water stress phase, trait data measured directly following the recovery phase for plants under water stress was analyzed with a one-way ANOVA with Genotype as the main effect.
Results
Response of S. melongena to water stress (prior to recovery watering)
Among the 25 traits measured, all but four were negatively impacted by the water stress treatment (LSM analysis, p< 0.0031, Table 1). The four traits which did not significantly change in response to water stress were all fine root traits: basal fine root mass, total fine root volume, fine root tissue density, and specific fine root length. Importantly, water stress significantly limited growth rate and photosynthetic activity relative to well-watered controls among all four genotypes. The mean maximum An rate of well-watered plants was 22.19 ± 1.28 μmol CO2 m-2 s-1 whereas plants under water stress attained maximum An rates of only 5.92 ± 1.16 μmol CO2 m-2 s-1 (LSM analysis, p < 0.0001, Table 1). Aboveground biomass (leaf dry weight, shoot biomass, and stem dry mass) was significantly reduced in plants under water stress as stem dry mass was 11.47 ± 1.26 g for well-watered plants compared to the much lower mass of plants under water stress, 7.40 ± 1.26 g (LSM analysis, p < 0.0001, S1 Table in S1 File). Belowground traits also followed this trend, with coarse and fine root biomass being significantly reduced in plants under water stress. Notably, however, plants under water stress had a significantly greater root:shoot ratio and RMF than the controls (15 and 20% greater; LSM analysis, p < 0.0002, Table 1). Morphological changes in SLA contributed to the reduction in leaf area for plants under water stress (SLA = 0.013 ± 0.0006 m2 g-1) relative to well-watered plants (SLA = 0.018 ± 0.0006 m2 g-1, LSM analysis, p < 0.0001, Table 1). Finally, hydraulic status of plants under water stress also declined regardless of genotype relative to control plants; leaf water potential became much more negative (-2.299 ± 0.119 vs. -0.814 ± 0.119 MPa, p<0.0001).
Responses among genotypes of S. melongena to water stress (prior to recovery)
For many traits, there were clear varietal differences. Across all organs, PHL 2778 had the highest biomass at the end of the water stress period, while PHL 2789 tended to have the lowest. Mara and PHL 4841 were generally intermediate in growth and biomass. Variation in stem biomass among genotypes typified the pattern in which PHL 2778 accumulated higher biomass (11.11 ± 0.39 g) than Mara (6.43 ± 0.39 g), PHL 4841 (6.34 ± 0.39 g), and PHL 2789 (5.56 ± 0.39 g, ANOVA, p< 0.0001, Table 2, Fig 1, left panel). Notably, the genotype with the highest total biomass also allocated the most to root biomass: PHL 2778 had the highest RMF, PHL 4841 intermediate, and Mara and PHL 2789 the lowest (ANOVA, p = 0.001; Table 2, Fig 2).
The solid box contains the inner two quartiles, whereas whisker lines indicate 1.5 x IQR (3rd quartile minus the 1st quartile) from the box. The white horizontal line is the mean.
The solid box contains the inner two quartiles, whereas whisker lines indicate 1.5 x IQR (3rd quartile minus the 1st quartile) from the box. The white horizontal line is the mean.
An opposite trend was observed with leaf morphology and plant hydraulic status, where the smaller plants of PHL 2789 had higher leaf area per unit dry weight (SLA), potentially due to better hydraulic status, supporting higher leaf area expansion than the larger plants of PHL 2778 (Table 2, Fig 3). PHL 2789 displayed the lowest hydraulic stress as indicated by a higher ΨL under water stress (-2.01 ± 0.06 MPa), followed by Mara (-2.07 ± 0.06 MPa), PHL 2778 (-2.47 ± 0.06 MPa) and PHL 4841 (-2.65 ± 0.06 MPa) (ANOVA, p< 0.0001, Table 2, Fig 4, left panel). PCA illustrates the broader differences between genotypes and their strategies in response to water stress (Fig 5). Greater biomass and proportional root allocation were observed in PHL 2278 and associated with more negative ΨL, reduced electron transport through photosystem II (low Fv/Fm values), smaller, thicker leaves (low SLA) and shorter SRLFineRts. On the other hand, PHL 2789 and Mara maintained better hydraulic status and electron transport, with larger, thinner leaves, but produced less root and shoot biomass.
The solid box contains the inner two quartiles, whereas whisker lines indicate 1.5 x IQR (3rd quartile minus the 1st quartile) from the box. The white horizontal line is the mean.
The solid box contains the inner two quartiles, whereas whisker lines indicate 1.5 x IQR (3rd quartile minus the 1st quartile) from the box. The white horizontal line is the mean.
Loading arrows indicate the magnitude and direction of trait variation (9 traits) underlying the sample distribution. Ellipses indicate groupings by genotype.
Other traits that demonstrated significant varietal differences were root:shoot ratio, LMA, senesced leaf dry mass, LAR, total shoot mass, total root mass, fine root mass, coarse root mass, and total fine root surface area (ANOVA, Table 2).
The remaining traits responded to water stress on a species level with no significant differences among genotypes in the water stress response (ANOVA, Table 2). These traits included all gas exchange/photosynthetic traits (An, Fv/Fm ratio, gs, transpiration, WUEi) as well as some plant morphological and biomass allocation traits (e.g., total leaf area, green leaf mass, basal fine root mass, SRLFineRts, total fine root volume, and total FRL:LA ratio).
Re-watering and recovery in water stressed S. melongena
Recovery via re-watering after water stress restored the following traits impacted by water stress back to values observed in control plants: physiological traits (Fv/Fm and ΨL, Fig 4, right panel), biometric traits (basal fine root mass and fine root tissue density), and allometric traits (LAR, RMF, SRLFineRts, FRL:LA ratio, and SLA, Fig 3, right panel). Full recovery for these traits was inferred from the lack of significant difference between mean values of control (well-watered) and re-watered plants that underwent water stress or, in the case of Fv/Fm, increase of trait values in re-watered plants beyond that of control plants (LSM analysis, Table 3).
The following traits did not attain full recovery to control plant levels within the week after re-watering, as evidenced by their significant differences from control plants after recovery treatment (LSM analysis, Table 2): total shoot mass, stem mass, total leaf area, green leaf mass, senesced leaf mass, fine root mass, coarse root mass, total root mass, total fine root length, fine root surface area, and fine root volume (Fig 2, right panel). These traits were all a function of plant growth rate, which certainly increased in during recovery but not enough to match the final biomass values of the consistently watered control plants.
Elimination of genotypic differences by recovery watering
Most traits that exhibited water stress-induced genotypic differences maintained these significant differences after re-watering (ANOVA, S1 Table in S1 File). As examples, stem dry mass remained lowest in PHL 2789 (8.62 ± 0.82 g) and highest in PHL 2778 (13.37 ± 0.82 g, ANOVA, p = 0.0033, Fig 2, right panel); and SLA remained highest in PHL 2789 (0.0170 ± 0.0006 m2 g-1) and lowest in PHL 2778 (0.0140 ± 0.0006 m2 g-1, ANOVA, p = 0.0237, Fig 3, right panel). Other traits include LAR, senesced leaf mass, total shoot mass, fine root mass, coarse root mass, total root mass, total fine root mass, RMF, and root:shoot ratio.
For two traits, however, recovery after re-watering eliminated the genotype-specific differences observed during water stress: neither leaf water potential (Fig 4, right panel) nor total fine root surface area displayed significant differences between genotypes after re-watering (ANOVA, S1 Table in S1 File).
Linking greenhouse observations to field-grown plants
Stem dry mass, total shoot dry mass, root dry mass, and LMA significantly correlated with fresh fruit mass in water stressed eggplant in field-grown trials (Pearson Correlation Coefficient, S2 Table in S1 File). Using the strongest relationship (root dry mass, R2 = 0.30, p = 0.0053) to project fresh fruit production from the greenhouse-grown plants in this study, we found that water stressed plants would have produced, at best, 83% of the fresh fruit yield of irrigated plants. PHL 2778 and Mara were projected to produce more fruit biomass under drought relative to their irrigated control (246.58 g/plant or 83% of irrigated controls and 154.26 g/plant or 77% of irrigated controls, respectively) than PHL 4841 and PHL 2789 (producing only 173.22 g/plant or 75% of irrigated controls and 135.09 g/plant or 67% of irrigated controls, respectively).
Discussion
The four S. melongena genotypes in this study appeared to use two different strategies in response to water stress during the vegetative phase. The first strategy is that of being a “water spender”, similar to the “early vigor” and “drought avoidance” models in agronomy and ecology [11, 12, 16, 25]. One such genotype was PHL 2778, which produced large, fast growing, plants with extensive root systems that quickly accessed water but experienced rapidly declining photosynthetic function and water status as the water supply was depleted. On the other hand, genotypes Mara and PH: 2789 were “water savers”, displaying hallmarks of classic drought tolerance [12, 16]. Mara and PHL 2789 were slow growing, smaller plants that maintained safer water use thresholds by using their available soil water more conservatively (Fig 5). Field data confirmed that larger S. melongena plants under water stress tended to have lower SLA (the inverse of specific leaf weight, which has a significant relationship with stem dry mass, S2 Table in S1 File), which could be an adaptive trait to reduce transpirational water loss. Interestingly, gas exchange parameters, and, thus, WUEi, did not differ among the genotypes, indicating that while smaller, the leaves of PHL 2778 did not have more sensitive gs thresholds or lower An. Ultimately, genotypic differences in the ontogeny of root expansion and depletion of pot water, as indicated by RMF and ΨL, were the main traits underpinning the productivity of different genotypes under water stress.
In general, increased RMF is a common allometric adjustment to drought, and may be driven by biomechanical differences in root turgor vs. shoot turgor under water deficit [63]. Root traits of S. melongena found here upheld previous observations that increased rooting investment largely results in increased productivity during drought (Table 2, Fig 2) [31, 64]. In oats, drought tolerant genotypes had increased root length, root surface area and length of fine root hairs [40, 65]. Additionally, deep root systems have been found to be associated with drought tolerance in common bean and with hydraulic efficiency thereby contributing to the fitness of monocots under drought [27, 66]. Plant strategies of substantial biomass allocation to root structures are generally expected to generate a tradeoff of decreased whole plant biomass accumulation. Conversely, a strategy of greater proportional aboveground biomass would increase photosynthetic carbon assimilation and thus increase biomass production [67]. The lack of such a tradeoff here could be due to slow-growing genotypes having additional sensitivities under water stress that limited their growth, such as limited cell expansion or starch accumulation (since WUEi did not differ among genotypes).
Remarkably, increased rooting investment in S. melongena genotypes did not translate to superior hydraulic status. In fact, results were the opposite: ΨL was more negative and LAR was smaller in PHL 2778 and PHL 4841, the genotypes with greater RMF, root mass, and total biomass (Table 2, Fig 5). These trait patterns may have been a result of plant confinement in pots, given that greenhouse-grown plants possessed finite pot space and limited irrigation was supplied daily. A more extensive root system may have allowed PHL 2778 and PHL 4841 to use up the water more quickly in their pots leading to more negative ΨL. It is possible that in the field, the deeper roots PHL 2778 and PHL 4841 would access additional water from deep soils and maintain ΨL and leaf area closer to those of the smaller, more hydraulically conservative genotypes (Mara and PHL 2789).
When drought is episodic, plant response to re-watering and plant performance under water stress are equally important for achieving acceptable yields [34]. This study identifies traits that remain impaired beyond the water stress exposure period and traits that recovered quickly in response to re-watering in S. melongena. After recovery via re-watering, genotypic differences were largely maintained in biomass-related traits (total and specific LA, dry stem weight, and root biomass). Physiological traits having to do with photosynthesis, hydraulic status, and labile biomass such as leaves and fine roots, however, lost their initial water stress-induced genotypic differences (Fv/Fm, RMF, fine root length, fine root SA, and ΨL) (Fig 4). These traits showed a high degree of short-term (1 week) recovery potential, suggesting the activation of a broad suite of key recovery traits such as membrane and photosystem repair, embolism recovery, and fine root flushing [8, 31, 68, 69]. Given early enough drought and a longer growing season remaining for S. melongena after resumption of adequate precipitation in the Philippines, there may be potential for more substantial biomass recovery across genotypes beyond that observed in this study.
Given that PHL 2778 achieved significantly greater biomass than the other genotypes under water stress in this study, and that biomass (of the leaves, shoot, and/or root) significantly correlated with fresh fruit production, it is clear that drought avoidance is the most promising strategy for moderate drought stress in eggplant. Breeding for rapid growth and, perhaps, early flowering will likely increase yields in S. melongena whether or not the plants experience drought in during the growing season. The other three genotypes, particularly Mara and PHL 2789, fall into the category of plants whose drought tolerance comes at a cost to biomass and productivity [16, 22, 70, 71]. They are thus less desirable for integrating into current crop genotypes. It should be noted, however, that these less productive, drought tolerant genotypes may outperform large drought avoiders like PHL 2778 when grown under more severe or repetitive drought, as it has been observed in maize [22]. Under current model of projected episodic drought in the Philippines, however, fast-growing genotypes such as PHL 2778 may provide the greatest agricultural benefit. The strategy of early high root investment and fast biomass growth compensates for rapidly declining hydraulic status to produce greater fruit yield, with strong potential for additional yield recovery after re-watering.
In conclusion, the identification of key trait responses to water stress will be useful for guiding selection of parental lines for varietal improvement, providing a clear desired trait outcome during the hybrid screening process. In this case, evaluation of water stress strategies among a set of demonstrated drought tolerant genotypes showed a range of whole-plant strategies, from small but tolerant water-savers to large, productive, but profligate water spenders. Given that these trait responses were evaluated in a greenhouse study, it will be important to verify the consistency of these strategies under field conditions where linkages to yield can also be confirmed, a necessary step to validate them as targets for breeding efforts [5].
Supporting information
S1 File. Contains supporting S1 and S2 Tables.
https://doi.org/10.1371/journal.pone.0256342.s001
(PDF)
Acknowledgments
We wish to thank Ryan Barton, Josh Wentz, Amy Nielsen, Deb Newman, Sean Gleason, Julie Bushey, Courtney Jahn, Kevin Yemoto, and Huihui Zhang for assistance with experimental set-up, data collection and plant processing.
References
- 1. Field CB, Barros V, Stocker TF, Qin D, Dokken DJ. Managing the risks of extreme events and disasters to advance climate change adaptation. A special report of working groups i and ii of the intergovernmental panal on climate change. IPCC. 2012.
- 2.
von Mogel HK. Interactions key to beating future droughts. CSA News. 2013.
- 3. Condon AG. Drying times: Plant traits to improve crop water use efficiency and yield. J Exp Bot. 2020;71: 2239–2252. pmid:31912130
- 4. Ferguson JN. Climate change and abiotic stress mechanisms in plants. 2019; 1–17.
- 5. Passioura JB. Phenotyping for drought tolerance in grain crops: When is it useful to breeders? Funct Plant Biol. 2012;39: 851–859. pmid:32480835
- 6. Volaire F. A unified framework of plant adaptive strategies to drought: Crossing scales and disciplines. Glob Chang Biol. 2018;24: 2929–2938. pmid:29350812
- 7. Khadka K, Earl HJ, Raizada MN, Navabi A. A Physio-Morphological Trait-Based Approach for Breeding Drought Tolerant Wheat. Front Plant Sci. 2020;11: 1–26. pmid:32117356
- 8. Ocheltree TW, Gleason SM, Cao K-F, Jiang G-F. Loss and recovery of leaf hydraulic conductance: Root pressure, embolism, and extra-xylary resistance. J Plant Hydraul. 2020;7: 1–16.
- 9. Ferguson JN, Humphry M, Lawson T, Brendel O, Bechtold U. Natural variation of life-history traits, water use, and drought responses in Arabidopsis. Plant Direct. 2018;2: e00035. pmid:31245683
- 10. Lof H. Water Use Efficiency and Competition Between Arid Zone Annuals, Especially the Grasses Phalaris minor and Hordeum murinum. Agric Res Reports. 1976; viii + 109 pp.
- 11. Ackerly D. Functional strategies of chaparral shrubs in relation to seasonal water deficit and disturbance. Ecol Monogr. 2004;74: 25–44.
- 12.
De Micco V, Aronne G. Morpho-anatomical traits for plant adaptation to drought. In: Aroca R, editor. Plant Responses to Drought Stress: From Morphological to Molecular Features. Springer-Verlag; 2012. p. 37. https://doi.org/10.1007/978-3-642-32653-0
- 13. Zwicke M, Picon-Cochard C, Morvan-Bertrand A, Prud’homme MP, Volaire F. What functional strategies drive drought survival and recovery of perennial species from upland grassland? Ann Bot. 2015;116: 1001–1015. pmid:25851134
- 14. Bristiel P, Roumet C, Violle C, Volaire F. Coping with drought: root trait variability within the perennial grass Dactylis glomerata captures a trade-off between dehydration avoidance and dehydration tolerance. Plant Soil. 2019;434: 327–342.
- 15. Basu S, Ramegowda V, Kumar A, Pereira A. Plant adaptation to drought stress [version 1; referees: 3 approved]. F1000Research. 2016;5: 1–10. pmid:27441087
- 16. Chater CC, Caine RS, Tomek M, Wallace S, Kamisugi Y, Cuming AC, et al. Origin and function of stomata in the moss Physcomitrella patens. Nat Plants. 2016;2: 16179. pmid:27892923
- 17. Condon AG, Richards RA, Rebetzke GJ, Farquhar GD. Breeding for high water-use efficiency. J Exp Bot. 2004;55: 2447–2460. pmid:15475373
- 18. Caine RS, Yin X, Sloan J, Harrison EL, Mohammed U, Fulton T, et al. Rice with reduced stomatal density conserves water and has improved drought tolerance under future climate conditions. New Phytol. 2019;221: 371–384. pmid:30043395
- 19. Gilbert ME, Zwieniecki MA, Holbrook NM. Independent variation in photosynthetic capacity and stomatal conductance leads to differences in intrinsic water use efficiency in 11 soybean genotypes before and during mild drought. J Exp Bot. 2011;62: 2875–2887. pmid:21339387
- 20. Khan HUR, Link W, Hocking TJ, Stoddard FL. Evaluation of physiological traits for improving drought tolerance in faba bean (Vicia faba L.). Plant Soil. 2007;292: 205–217.
- 21. Quarrie SA, Jones HG. Genotypic Variation in Leaf Water Potential, Stomatal Conductance and Abscisic Acid Concentration in Spring Wheat Subjected to Artificial Drought Stress. Ann Bot. 1979;44: 323–332.
- 22. Collins NC, Tardieu F, Tuberosa R. Quantitative trait loci and crop performance under abiotic stress: Where do we stand? Plant Physiol. 2008;147: 469–486. pmid:18524878
- 23. Bartels D, Sunkar R. Drought and salt tolerance in plants. CRC Crit Rev Plant Sci. 2005;24: 23–58.
- 24. Subbarao G V., Johansen C, Slinkard AE, Nageswara Rao RC, Saxena NP, Chauhan YS. Strategies For Improving Drought Resistance In Grain Legumes. CRC Crit Rev Plant Sci. 1995;14: 469–523.
- 25.
Turner NC, Nicolas ME. Early vigour: a yield-positive characteristic for wheat in drought-prone Mediterranean-type environments. In: Behl D.P. Singh GPL, editor. Crop Improvement for Stress Tolerance. New Delhi; 1998. pp. 47–62.
- 26. Whan BR, Carlton GP, Anderson WK. Potential for increasing early vigour and total biomass in spring wheat. I. Identification of genetic improvements. Aust J Agric Res. 1991;42: 347–361.
- 27. Gleason SM, Cooper M, Wiggans DR, Bliss CA, Romay MC, Gore MA, et al. Stomatal conductance, xylem water transport, and root traits underpin improved performance under drought and well-watered conditions across a diverse panel of maize inbred lines. F Crop Res. 2019;234: 119–128.
- 28. Slafer GA, Araus JL, Royo C, García Del Moral LF. Promising eco-physiological traits for genetic improvement of cereal yields in Mediterranean environments. Ann Appl Biol. 2005;146: 61–70.
- 29. Lafitte HR, Champoux MC, McLaren G, O’Toole JC. Rice root morphological traits are related to isozyme group and adaptation. F Crop Res. 2001;71: 57–70.
- 30. Kato Y, Abe J, Kamoshita A, Yamagishi J. Genotypic variation in root growth angle in rice (Oryza sativa L.) and its association with deep root development in upland fields with different water regimes. Plant Soil. 2006;287: 117–129.
- 31. Comas LH, Becker SR, Cruz VM V., Byrne PF, Dierig DA. Root traits contributing to plant productivity under drought. Front Plant Sci. 2013;4: 1–16. pmid:23346092
- 32. Shipley B, Meziane D. The balanced-growth hypothesis and the allometry of leaf and root biomass allocation. Funct Ecol. 2002;16: 326–331.
- 33. Kashiwagi J, Krishnamurthy L, Crouch JH, Serraj R. Variability of root length density and its contributions to seed yield in chickpea (Cicer arietinum L.) under terminal drought stress. F Crop Res. 2006;95: 171–181.
- 34.
Sperry JS, Stiller V, Hacke U. Soil Water Uptake and Water Transport Through Root Systems. 3rd ed. In: Waisel Y, Eshel A, Kafkafi U, editors. Plant Roots. 3rd ed. New York, NY: Marcel Dekker, Inc.; 2002. pp. 663–681. https://doi.org/10.1201/9780203909423.ch38
- 35. Lo Gullo MA, Nardini A, Salleo S, Tyree MT. Changes in root hydraulic conductance (K(R)) of Olea oleaster seedlings following drought stress and irrigation. New Phytol. 1998;140: 25–31.
- 36. Sadras VO, Rodriguez D. The limit to wheat water-use efficiency in Eastern Australia. II. Influence of rainfall patterns. Aust J Agric Res. 2007;58: 657–669.
- 37. Yousfi N, Saïdi I, Slama I, Abdelly C. Phenology, leaf gas exchange, growth and seed yield in Medicago polymorpha L. populations affected by water deficit and subsequent recovery. Flora Morphol Distrib Funct Ecol Plants. 2015;214: 50–60.
- 38. Chen D, Wang S, Cao B, Cao D, Leng G, Li H, et al. Genotypic variation in growth and physiological response to drought stress and re-watering reveals the critical role of recovery in drought adaptation in maize seedlings. Front Plant Sci. 2016;6: 1–15. pmid:26793218
- 39. Gleason SM, Wiggans DR, Bliss CA, Comas LH, Cooper M, DeJonge KC, et al. Coordinated decline in photosynthesis and hydraulic conductance during drought stress in Zea mays. Flora. 2017;227: 1–9.
- 40. Canales FJ, Nagel KA, Müller C, Rispail N, Prats E. Deciphering Root Architectural Traits Involved to Cope With Water Deficit in Oat. Front Plant Sci. 2019;10: 1–15. pmid:30723482
- 41. Yusuf AA, Francisco H. Climate Change Vulnerability Mapping for Southeast Asia Vulnerability Mapping for Southeast Asia. East. 2009;181: 1–19.
- 42. Jose AM, Cruz NA. Climate change impacts and responses in the Philippines: Water resources. Clim Res. 1999;12: 77–84.
- 43. Thirumalai K, DiNezio PN, Okumura Y, Deser C. Extreme temperatures in Southeast Asia caused by El Ninõ and worsened by global warming. Nat Commun. 2017;8: 1–8.
- 44. Hughes RDLP and J. Improving Vegetable Productivity in a Variable Changing Climate. An Open Access J Publshed by ICRISAT. 2007;4: 1–22. ejournal.icrisat.org
- 45. Bolaños J, Edmeades GO, Martinez L. Eight cycles of selection for drought tolerance in lowland tropical maize. III. Responses in drought-adaptive physiological and morphological traits. F Crop Res. 1993;31: 269–286.
- 46. Edmeades GO, Bolaños J, Chapman SC, Lafitte HR, Bänziger M. Selection improves drought tolerance in tropical maize populations: II. Direct and correlated responses among secondary traits. Crop Sci. 1999;39: 1315–1324.
- 47.
Opeña R, Chen J, Kuo C, Chen H. Genetic and Physiological Aspects of Tropical Adaptation in Tomato. Adaptation of Food Crops to Temperature and Water Stress. Shanhua, Taiwan: Asian Vegetable Research and Development Center; 1992. pp. 257–270.
- 48. Upadhyaya HD, Salimath PM, Gowda CLL, Singh S. New early-maturing germplasm lines for utilization in chickpea improvement. Euphytica. 2007;157: 195–208.
- 49. Ashraf M. Inducing drought tolerance in plants: Recent advances. Biotechnol Adv. 2010;28: 169–183. pmid:19914371
- 50. Ullah I, Mehboob-ur-Rahman , Ashraf M, Zafar Y. Genotypic variation for drought tolerance in cotton (Gossypium hirsutum L.): Leaf gas exchange and productivity. Flora Morphol Distrib Funct Ecol Plants. 2008;203: 105–115.
- 51. Guan XK, Song L, Wang TC, Turner NC, Li FM. Effect of Drought on the Gas Exchange, Chlorophyll Fluorescence and Yield of Six Different-Era Spring Wheat Cultivars. J Agron Crop Sci. 2015;201: 253–266.
- 52. Zhang H, Gong G, Guo S, Ren Y, Xu Y, Ling KS. Screening the USDA watermelon germplasm collection for drought tolerance at the seedling stage. HortScience. 2011;46: 1245–1248.
- 53. Lewis Ivey ML, McSpadden Gardener BB, Opina N, Miller SA. Diversity of Ralstonia solanacearum infecting eggplant in the Philippines. Phytopathology. 2007;97: 1467–1475. pmid:18943517
- 54.
Chupungco AR, Elazegui DD, Nguyen MR. The Eggplant Industry in the Philippines: Seed System, Production, and Marketing. Socioeconomic Impacts of Bt Eggplant: Ex-ante Case Studies in the Philippines. 2014. www.isaaa.org/resources/publications/socioeconomics_impacts_of_bt_eggplant/download/Socioeconomics Impacts of Bt Eggplant.pdf
- 55. Sarker BC, Hara M. Periodic Drought Stress Effect on Evapotranspiration, Root Hydraulic Conductance and Fruit Yield Efficiency of Eggplant. Asian J Plant Sci. 2004;3: 132–139.
- 56. Kirnak H, Kaya C, Tas I, Higgs D. The influence of water deficit on vegetative growth, physiology, fruit yield and quality in eggplants. 2001;27: 34–46.
- 57. Plazas M, Nguyen HT, González-Orenga S, Fita A, Vicente O, Prohens J, et al. Comparative analysis of the responses to water stress in eggplant (Solanum melongena) cultivars. Plant Physiol Biochem. 2019;143: 72–82. pmid:31491702
- 58. Amiri Rodan M, Hassandokht MR, Sadeghzadeh-Ahari D, Mousavi A. Mitigation of drought stress in eggplant by date straw and plastic mulches. J Saudi Soc Agric Sci. 2020;19: 492–498.
- 59. Sarker BC, Hara M, Uemura M. Proline synthesis, physiological responses and biomass yield of eggplants during and after repetitive soil moisture stress. Sci Hortic (Amsterdam). 2005;103: 387–402.
- 60.
Delfin EF. Development of Molecular Markers for the Assessment of Moisture Stress Tolerance in Eggplant Terminal Report. Quezon City; 2015.
- 61. Mibei EK, Ambuko J, Giovannoni JJ, Onyango AN, Owino WO. Carotenoid profiling of the leaves of selected African eggplant accessions subjected to drought stress. Food Sci Nutr. 2017;5: 113–122. pmid:28070322
- 62. Freschet GT, Violle C, Bourget MY, Scherer-Lorenzen M, Fort F. Allocation, morphology, physiology, architecture: the multiple facets of plant above- and below-ground responses to resource stress. New Phytol. 2018;219: 1338–1352. pmid:29856482
- 63. Hsiao TC, Xu LK. Sensitivity of with of roots versus leaves to water stress: Biophysical analysis and relation to water. J Exp Bot. 2000;51: 1595–1616. pmid:11006310
- 64. Awad W, Byrne PF, Reid SD, Comas LH, Haley SD. Great Plains winter wheat varies for root length and diameter under drought stress. Agron J. 2018;110: 226–235.
- 65. Polania J, Poschenrieder C, Rao I, Beebe S. Root traits and their potential links to plant ideotypes to improve drought resistance in common bean. Theor Exp Plant Physiol. 2017;29: 143–154. pmid:33552846
- 66. Ocheltree TW, Mueller KM, Chesus K, LeCain DR, Kray JA, Blumenthal DM. Identification of suites of traits that explains drought resistance and phenological patterns of plants in a semi-arid grassland community. Oecologia. 2020;192: 55–66. pmid:31932921
- 67. Bloom AJ, Chapin FS, Mooney HA. Resource limitation in plants—an economic analogy. Annu Rev Ecol Syst Vol 16. 1985; 363–392.
- 68. Trifilò P, Raimondo F, Lo Gullo M a, Barbera PM, Salleo S, Nardini A. Relax and refill: xylem rehydration prior to hydraulic measurements favours embolism repair in stems and generates artificially low PLC values. Plant, Cell Environ. 2014;37: 2491–9. pmid:24588546
- 69. Correia MJ, Osório ML, Osório J, Barrote I, Martins M, David MM. Influence of transient shade periods on the effects of drought on photosynthesis, carbohydrate accumulation and lipid peroxidation in sunflower leaves. Environ Exp Bot. 2006;58: 75–84.
- 70. Parry MAJ, Flexas J, Medrano H. Prospects for crop production under drought: Research priorities and future directions. Ann Appl Biol. 2005;147: 211–226.
- 71. Manavalan LP, Guttikonda SK, Phan Tran LS, Nguyen HT. Physiological and molecular approaches to improve drought resistance in soybean. Plant Cell Physiol. 2009;50: 1260–1276. pmid:19546148