Metabolomic Profiling of Drought-Tolerant and Susceptible Peanut (Arachis hypogaea L.) Genotypes in Response to Drought Stress.

Peanut is frequently constrained by extreme environmental conditions such as drought. To reveal the involvement of metabolites, TAG 24 (drought-tolerant) and JL 24 (drought-sensitive) peanut genotypes were investigated under control and 20% PEG 6000-mediated water scarcity conditions at the seedling stage. Samples were analyzed by gas chromatography-mass spectrometry (GC-MS) to identify untargeted metabolites and targeted metabolites, i.e., polyamines and polyphenols by high-performance liquid chromatography (HPLC) and ultrahigh-performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS), respectively. The principal component analysis (PCA), partial least-squares discriminant analysis (PLS-DA), heat map, and cluster analysis were applied to the metabolomics data obtained by the GC-MS technique to determine the important metabolites for drought tolerance. Among 46 resulting metabolites, pentitol, phytol, xylonic acid, d-xylopyranose, stearic acid, and d-ribose were important drought-responsive metabolites. Agmatine and cadaverine were present in TAG 24 leaves and roots, respectively, during water-deficit conditions and believed to be the potential polyamines for drought tolerance. Polyphenols such as syringic acid and vanillic acid were produced more in the leaves of TAG 24, while catechin production was high in JL 24 during stress conditions. Seven metabolic pathways, namely, galactose metabolism, starch and sucrose metabolism, fructose and mannose metabolism, pentose and glucuronate interconversion, propanoate metabolism, amino sugar and nucleotide sugar metabolism, and biosynthesis of unsaturated fatty acids were significantly affected by water-deficit conditions. This study provides valuable information about the metabolic response of peanut to drought stress and metabolites identified, which encourages further study by transcriptome and proteomics to improve drought tolerance in peanut.

Introduction

Peanut (Arachis hypogaea L.) is an important oilseed and cash crop cultivated in semiarid zones of the world, and it is mainly cultivated in Asia, Africa, and America.[1] The seed is more valued because of unsaturated edible oil (48–50%), easily digestible protein (26–28%), half of the essential vitamins, and one-third of the essential minerals.[2] About 80% of the world groundnut production comes from seasonally rainfed areas in the subtropics, where the climate is characterized by low and erratic rainfall. According to a recent estimate, global peanut productivity incurred an annual loss of approximately 6 million tons due to drought alone among all abiotic stress factors.[3−5] Thus, it is essential to reveal the mechanisms of drought tolerance and identify drought-resistant peanut germplasms.[6]

Metabolomics research in plant systems is progressing; it measures all or a set of metabolites present in a specified sample during a particular time. Overall, the metabolomes of higher plants are estimated to consist of more than 100 000 primary and secondary metabolites out of which roughly 10% have been recognized to date.[7] The quantitative and qualitative compositions of plant metabolomes reflect their responses to biotic and abiotic stimuli, genome, and physiological status, thus serving as a connecting link between genotypes and phenotypes. It makes a significant contribution to the research of stress biology by recognizing various compounds such as byproducts of stress metabolism, stress signal transduction molecules, and molecules that are part of the plant acclimation process.[8,9] Metabolites can be regarded as the ultimate response to environmental changes.[10,11] Several cellular metabolites are altered during drought stress, such as soluble sugars, organic acids, phenolics, amino acids, fatty acids, nucleotides, peptides, cofactors, and secondary metabolites.[12] Many of these metabolites are vital components of the plant’s defense system.[13] Polyamines and phenolic compounds (phenolic acids and flavonoids) are substantial groups of plants secondary metabolites that impart tolerance and are described as a new kind of biostimulants under environmental stress, especially drought stress conditions.[14,15]

The current study was aimed to compare metabolic changes in leaves and roots of drought-tolerant and -sensitive peanut genotypes when subjected to drought stress at the seedling stage. For the imposition of drought in vitro, poly(ethylene glycol)-6000 used. Studies reported that PEG induces significant water stress in plants without causing toxic effects and any physiological damage.[16] For putative identification of drought-specific metabolites, we employed a gas chromatography–mass spectrometry (GC–MS)-based untargeted metabolomics approach. A targeted metabolomics approach was used to evaluate polyamines and polyphenol through high-performance liquid chromatography (HPLC) and UPLC–MS/MS, respectively. The metabolic content of peanut genotypes was compared to reveal the effects of drought stress on the metabolomic level. These results provide insights into metabolites involved in the mechanisms of plant drought tolerance, which can eventually contribute to the future genetic and metabolomics studies of domesticated crops. To our best knowledge, this is the first time that a metabolic comparison has been made in cultivated peanut in leaf and root samples via GC–MS and LC–MS analyses.

Result and Discussion

Untargeted Metabolites

The current study carried out to understand the metabolic alteration in different parts (leaves and roots) of the plant at the seedling stage that could provide a more precise indication of stress tolerance in plants. In the present study, a total of 46 and 29 metabolites were accumulated in leaf and root extracts of peanut, respectively, as determined from the chromatogram. Using the NIST library, metabolites were identified as sugars (47%), sugar alcohols (13%), sugar acids (9%), fatty acids (9%), and others such as dicarboxylic acid, diterpene alcohol, organic acid, and sugar amine. The total number of metabolites produced in each sample are given in the Supporting Information (Table S1).

Heat map analysis of all metabolites in the leaf and root samples of TAG 24 under the water stress revealed a high accumulation of sugars such as mannose, d-ribose, d-xylopyranose, β-d-galactopyranoside, α-d-glucopyranose (Figure ) and d-ribose, 2-deoxyribose, galactose oxime, β-d-galactopyranose, and l-manopyranose, respectively. Metabolites such as pentitol (sugar alcohol), 3,7,11,15-tetramethyl-2-hexadecen-1-ol (diterpene alcohol also known as phytol), saturated fatty acids such as stearic acid, xylonic acid (sugar acid), and myo-inositol (sugar amine) were detected only in the leaf sample and pentadecanoic (fatty acid) and galactosoxime (sugar amine) were detected in roots of TAG 24 under given stress. In the case of water stress-sensitive genotype JL 24, sugars such as d-fructose and d-turanose, organic acids such as 2,3,4-trihydroxybutyric acid and malic acid, and dicarboxylic acids such as succinic acid and 2 butenedoic acids were present in leaf samples, while in roots, 2-deoxyribose and galactofuranose (sugars), d-mannitol (sugar alcohol), myo-inositol and glucose oxime (sugar amines), and 8,11-octadecadienoic acid (fatty acid) were accumulated under water stress. In control of leaf samples, 2-deoxy-galactopyranose, d-galactose, and d-glucose were present in TAG 24 and JL 24, while in JL 24, propanoic acid, galactaric acid, and α-d-galactoside were also present as additional metabolites. In control of root samples, metabolites such as arabinitol, d-mannitol, maltose, d-turanose, d-xylopyranose, and ribonic acid were present in TAG 24, while palmitic acid, pentadecanoic acid, sorbitol, β-d-mannopyranoside, and l-mannopyranose were present in JL 24.

Figure 1

Heat map analysis showing abundance of metabolites during control and drought stress in leaves and roots of the tolerant (TAG 24) and susceptible (JL 24) genotypes, where JCL = JL 24 leaf (control), TCL = TAG 24 leaf (control), JSL = JL 24 leaf (stress), TSL = TAG 24 leaf (stress), JCR = JL 24 root (control), TCR = TAG 24 root (control), JSR = JL 24 root (stress), TSR = TAG 24 root (stress).

Different metabolites of tolerant and sensitive genotypes were also determined using variable importance in projection (VIP) measure of partial least-squares discriminant analysis (PLS-DA) and metabolites with VIP score > 1 were explained.[17] PLS-DA is a chemometric method used to optimize the separation between different groups.[18] The VIP score (Figure a) of leaf samples showed higher intensity of stress-specific metabolites, such as 3,7,11,15-tetramethyl-2-hexadecen-1-ol, pentitol, d-ribose, and d-xylopyranose in the tolerant genotype and 3,7,11,15-tetramethyl-2-hexadecen-1-ol (phytol), d-turanose, 2-O-glycerol, xylulose, and galacteric acid in the sensitive genotype. In the control sample, metabolites were d-glucose, pentitol, melibiose, xylulose, and α-d-manopyranose in TAG 24, while d-ribose, 2-O-glycerol, galacteric acid, and α-d-mannopyranose were present in higher concentrations in the JL 24 genotype. In root samples (Figure b), d-ribose and α-d-glucopyranose were observed in TAG 24, while in JL 24, palmitic acid and myo-inositol were produced under stress conditions. In the control sample, d-ribose and myo-inositol were observed in TAG 24 and palmitic acid was observed in the JL 24 genotype.

Figure 2

(a) Abundant metabolites identified in the leaf sample using partial least-squares discriminant analysis (PLS-DA) using variable importance in the projection (VIP) score in control and drought stress, where JCL = JL 24 leaf (control), TCL = TAG 24 leaf (control), JSL = JL 24 leaf (stress), and TSL = TAG 24 leaf (stress). (b) Abundant metabolites identified in the root sample using partial least-squares discriminate analysis (PLS-DA) using variable importance in the projection (VIP) score in control and drought stress, where JCR = JL 24 root (control), TCR = TAG 24 root (control), JSR = JL 24 root (stress), and TSR = TAG 24 root (stress).

Based on dendrogram analysis (Figure ), metabolites of root and leaf samples of both the genotypes were scattered into two main clusters. Cluster-1 has majority of metabolites under two subclusters. Subcluster-1 has the majority of leaf metabolites, whereas subcluster-2 has root metabolites. TAG 24 root metabolites and sensitive JL 24 leaf metabolites share the same sub-subcluster-1, whereas sub-subcluster-2 has metabolites of the leaf of both varieties in the control condition.

Figure 3

Clustering pattern shown as the dendrogram of peanut genotypes in control and drought stress, where JCL = JL 24 leaf (control), TCL = TAG 24 leaf (control), JSL = JL 24 leaf (stress), TSL = TAG 24 leaf (stress), JCR = JL 24 root (control), TCR = TAG 24 root (control), JSR = JL 24 root (stress), and TSR = TAG 24 root (stress).

Sugars and their derivatives may have accumulated in response to stress and can function as osmolytes to maintain cell turgor and provide a hydration shell around proteins, thereby providing the first line of defense against further water loss and may assist to maintain a water balance in drought-tolerant plants.[19−22] Sugar also acts as a signaling molecule and helps to modulate the plant’s growth, development, and response to multiple stresses. Mannose was found to be accumulated in the leaf of TAG 24 in stress conditions; similar findings were reported earlier in the drought-tolerant wheat variety (JD17). Increased mannose under drought stress is mainly attributed to the enhanced hexokinase, implicating the improved capability of the drought-tolerant genotype in the biosynthesis of sugar and carbon storage.[23]

In the present study, we noted the accumulation of sugar alcohols such as pentitol, myo-inositol, and d-mannitol in the leaf sample of TAG 24 in response to water stress. Sugar alcohols are osmoprotectants and accumulated in different concentrations under water stress. The hydroxyl group of sugar alcohol can substitute the hydroxyl group of water during interaction with membrane lipids and proteins, maintaining their structure and properties under drought conditions.[24] TAG 24 showed an accumulation of pentitol in leaves as well as in roots when plants experienced low water potentials. The accumulation was higher in leaves than in roots that might lead to better growth and enhance the tolerance mechanism. The encouraging effects of mannitol in drought and salinity tolerance in wheat were demonstrated earlier.[25] The cyclic polyol (myo-inositol) was accumulated in the leaves of TAG 24 in drought exposed plants, which is supported by previous studies that myo-inositol imparts drought tolerance in various crops.[20,26−28]

Besides sugar alcohol, we observed the effect of drought on saturated fatty acids. The osmotic stress was positively correlated with an elevated level of stearic acid and pentadecanoic acid, indicating that membrane damage is related to the elevated levels of fatty acids. Similar findings were reported in leaf saps of the tolerant wheat genotype[20] and Flega (drought-tolerant oat cultivar) plants under drought. The increased concentration of pentadecanoic acid in the root of TAG 24 in response to water stress was in agreement with the previously reported study of drought tolerance.[29] The 3,7,11,15-tetramethyl-2-hexadecen-1-ol is an unsaturated long-chain fatty acid alcohol.[30] It was accumulated very high in TAG 24 leaves in stress conditions, efficiently scavenging the free radicals and imparting drought tolerance by protecting plants against oxidative stress.[31] Our results draw attention to palmitic acid, which shows a negative correlation with the drought stress response. Similar findings were also reported in safflower seeds (Carthamus tinctorius L).[32] In plants, fatty acid metabolic pathways play an important role in plant defense. Fatty acids and lipids are now recognized as more than just storage compounds or membrane structural components. Fatty acids also regulate processes such as growth and development and responses to biotic and abiotic stresses for acclimation.[33] It also reported that it preserves cell compartmentation, during drought stress.[34]

Targeted Metabolites

Polyphenols

Phenolic compounds (phenolic acids and flavonoids) are substantial groups of plant secondary metabolites and well known as a marker of biotic and abiotic stress tolerance. Phenolic compounds may exhibit antioxidant activity to plants by scavenging reactive oxygen species. Sixteen phenolics, cinnamic acid, caffeic acid, salicylic acid, gallic acid, ferulic acid, quercetin, catechol, chlorogenic acid, coumaric acid, syringic acid, kaempferol, vanillic acid, catechin, epicatechin, and epigallocatechin, were identified and quantified using UPLC–MS/MS (Table S2) in leaf and root samples of both the genotypes. Drought-induced production of endogenous phenolic compounds has been reported in peanut, wheat, maize, desert shrub.[15,35−37] It has been also reported earlier that polyphenols responsible for controlling the osmotic potential and proline metabolism provide tolerance to various abiotic stresses.[38,39]

A heat map analysis revealed that vanillic acid and syringic acid were accumulated in the leaf sample of TAG 24 in stress conditions (Figure ). However, vanillic acid was present in all four stressed samples, but the concentration was higher in TAG 24 than JL 24, whereas it was absent in the control sample. The higher accumulation of vanillic acid (hydroxybenzoic acid) in the drought-tolerant genotype as compared to the sensitive genotype implicates that vanillic acid could provide resistance against drought stress and thereby provide drought tolerance.[35,40] Cinnamic acid and caffeic acid were produced in leaf and root samples of TAG 24 in stress conditions and reported for tolerance in other crops.[41,42] It was reported that cinnamic acid helps to reduce lipid peroxidation and increases the activities of antioxidant enzymes in drought-stressed cucumber leaves.[43] Phenolics compounds release hydroxyl group (OH) hydrogen atom, hence oxidizing themselves and acting as an antioxidant. The antioxidative property of phenols is attributed to the presence of the number of hydroxyl groups;[44] caffeic acid has two hydroxyl groups, so it has a much higher antioxidant effect and may give enhanced tolerance to drought.

Figure 4

Heat map analysis of phenolics quantified during control and drought stress in leaf and root of the tolerant (TAG 24) and susceptible (JL 24) genotypes, where JCL = JL 24 leaf (Control), TCL = TAG 24 leaf (control), JSL = JL 24 leaf (stress), TSL = TAG 24 leaf (stress), JCR = JL 24 root (control), TCR = TAG 24 root (control), JSR = JL 24 root (stress), and TSR = TAG 24 root (stress).

In the root of TAG 24, salicylic acid (hydroxybenzoic acid) was accumulated predominantly and cinnamic acid and syringic acid were also present under stress conditions. Salicylic acid is considered as a plant growth regulator that improves plants’ response toward drought stress by maintaining a better rooting system.[40][45] In the control sample, gallic acid, quercetin, salicylic acid, cinnamic acid, and caffeic acid and epicatechin were present in the root sample of TAG 24 and JL 24, respectively. Other phenolics such as chlorogenic acid, epigallocatechin, catechol, and kaempferol were detected more in the control sample of leaves of TAG 24, whereas their concentrations were lower in stress conditions. Ferulic acid and coumaric acid were present in all samples of JL 24. Gallic acid was present in leaf and root samples of TAG 24 in the control, and their concentrations were lower in stress conditions. According to their presence and quantity synthesized, vanillic acid, cinnamic acid, syringic acid, and salicylic acid showed powerful involvement in the drought tolerance of TAG 24. Further experiments need to be performed to examine the mechanism of involvement in the drought resistance of peanut. Furthermore, the examination of derivatives including enzymes and proteins synthesizing these phenolic acids, particularly vanillic acid and syringic acid, requires to be investigated in detail.

Polyamines

Polyamines are considered as biostimulants of plants as they play an important role in plant growth and development and also in a reaction to environmental stress.[14] Five polyamines, i.e., putrescine, cadaverine, agmatine, spermidine, and spermine, were identified and quantified in leaf and root extracts under stress and control conditions of tolerant and sensitive peanut genotypes by HPLC.

The heat map analysis of polyamines suggested that the concentration of agmatine increased in TAG 24 and decreased in JL 24 under stress conditions (Figure ), indicating that this metabolite was a key polyamine produced in response to water stress. Its accumulation was 1.3 and 10 times higher in leaves and roots of TAG 24, respectively, under stress conditions as compared to its control. In contrast, agmatine decreased under stress conditions in JL 24. The concentration lowered 5 and 3 times than the control sample in leaves and roots, respectively (Table S3). The accumulation of agmatine in stress conditions indicated its crucial role in drought tolerance in peanut.[46,47] It is likely that under stress conditions, the concentration of agmatine is controlled by the conversion of agmatine into putrescine by agmatine deiminase. Second, arginine converts into N-carbamoyl putrescine and then into agmatine by arginine decarboxylase and N-carbamoyl-putrescine-amido-hydrolase, respectively. Agmatine is the main precursor of important polyamines found in living cells, i.e., putrescine, spermidine, and spermine, and is reported to impart osmotic protection to drought stress.[48−50]

Figure 5

Heat map analysis of polyamines quantified during control and drought stress in leaves and roots of the tolerant (TAG 24) and susceptible (JL 24) genotypes, where JCL = JL 24 leaf (control), TCL = TAG 24 leaf (control), JSL = JL 24 leaf (stress), TSL = TAG 24 leaf (stress), and JCR = JL 24 roots (control), TCR = TAG 24 root (control), JSR = JL 24 root (stress), and TSR = TAG 24 root (stress).

The heat map suggested that diamine cadaverine was produced predominantly in leaf samples of TAG 24 and JL 24 and root samples of TAG 24 in stress conditions. Its concentration was lower in control samples of both genotypes. High cadaverine accumulation was reported in other crops, such as in the oilseed rape leaves (45-fold) and in calli of the sensitive wheat cultivar in response to drought stress.[51,52] However, cadaverine concentration decreased in TAG 24 leaves in stress response. There is a dichotomy between cadaverine acting as a stress protectant or exacerbating stress damage. The concentration of putrescine was lower in stressed samples of TAG 24 compared to that in control samples of TAG 24, whereas spermidine and spermine were present in lower concentrations in both the samples of TAG 24 and JL 24 compared to the stressed sample. However, the concentration of spermine was relatively high in the roots of JL 24 under stress. Spermidine decreased in the root of both TAG 24 and JL 24 genotypes under stress, while spermine slightly increased in the roots of JL 24. Similar observation also reported earlier in leaves of rape seedlings subjected to drought stress.[51] These findings suggest a rise in putrescine biosynthesis and stimulation of polyamine oxidation reactions were monitored at the polyamine level during stress.

Another contrast results report stated that PEG 6000 treatment significantly increased the spermidine levels in leaves of Triticum aestivum drought-tolerant cultivar (Yumai No. 18 genotype) indicating that free-spermidine facilitated the osmotic stress tolerance of wheat seedlings.[53] Levels of putrescine and spermidine increased in drought stress in barley plants.[54] Putrescine is converted to spermidine by spermidine synthase (SPDS) and then to spermine by spermine synthase (SPMS). Spermidine and spermine are substrates of polyamine oxidases (PAOs), which catalyze the backconversion to putrescine.[19] Various abiotic stresses modulate polyamines levels, and their levels have been positively correlated with stress tolerance, so it is essential to check the concentration of polyamines to reveal tolerance.[19]

Metabolic Pathways

Drought affects important pathways and networks; analysis of relevant pathway enrichment and topology was performed using “Metabolic pathway analysis” (MetPA—a web-based tool) in MetaboAnalyst 4.0. Each metabolite involved in a pathway has unique biological functions. Metabolic pathway analysis was performed on altered known metabolites using Arabidopsis thaliana as the pathway library. The most impacted pathways having high statistical significance scores were annotated. A list of the affected pathway, the number of hit metabolites, and the false discovery rate (FDR) are mentioned in Table . Pathway topology analysis showed that seven pathways were significantly affected under drought condition (FDR < 1), viz., galactose metabolism, starch and sucrose metabolism, fructose and mannose metabolism, pentose and glucuronate interconversion, propanoate metabolism, amino sugar and nucleotide sugar metabolism, and biosynthesis of unsaturated fatty acids (Figure ).

Table 1

Detailed Results from the Metabolomic Pathway Analysisa

The name of pathways, total metabolites involved in these pathways, metabolites significantly accumulated in the present study (hits), and false discovery rate (FDR) are listed.

Figure 6

Metabolomic pathway as generated by the MetaboAnalyst software package. (All of the matched pathways are displayed as circles. The color of each circle is based on p-values; darker colors indicate more significant changes of metabolites in the corresponding pathway, whereas the size of the circle corresponds to the pathway impact score).

It was reported that genes for galactose metabolism, fructose and mannose metabolism, amino sugar and nucleotide sugar metabolism were upregulated in drought stress in drought-tolerant sesame to cope up with stress.[55] Sugars, such as GABA, galactose, fructose, and mannose, serve as metabolic precursors in many metabolic processes in plants.[56] In addition, starch, sucrose, and galactose metabolism was altered in Jatropha curcas drought-treated plants.[57] Also, pathways such as sucrose and starch metabolism and pentose and glucuronate interconversion were induced at the fiber initiation stage of cotton under drought stress, suggesting the regulation of sugar and energy metabolism to adapt to drought stress.[58] The succinic acid and propionic acids are key metabolites of the propanoate metabolism pathway; this pathway is induced in the tolerant Cicer arietinum L in drought stress.[59] Due to water stress, plants get adapted by alteration of fatty acid composition in membrane lipids. The polyunsaturation of fatty acids has proven concerning plant adaptation to abiotic stress. Enzymes like fatty acid desaturases are affected by water stress and hence increase unsaturated fatty acids.[60]

Conclusions

The present study demonstrated that 20% PEG 6000-treated leaves and roots of peanut genotypes differing in sensitivity to drought have a different mechanism of metabolite accumulation and regulation that is valuable for a better understanding of the overall abiotic stress tolerance mechanism. Overall, sugars, sugar alcohols, organic acids, and fatty acids are major groups of metabolites altered due to drought stress. Mannose, pentitol, myo-inositol, stearic acid, pentadecanoic acid, phytol, vanillic acid, cinnamic acid, syringic acid, salicylic acid, agmatine, and cadaverine showed upon water-deficit treatment by 20% PEG 6000, they can be considered as the principle drought stress-specific markers and osmoprotectants. Altered cardinal metabolites accumulated in leaves and roots of TAG 24 under stress conditions could be correlated with potential biochemical pathways and enzymes associated with them. A better depiction of the tolerance mechanism requires further investigation. A comparative study indicated that under drought conditions, levels of the total metabolites were found to be more pronounced in leaves than roots. The metabolomics approach will improve our insight into the underlying water deficiency mechanisms in the peanut and will become continually important in the future.

Materials and Methods

Plant Materials, Growth Conditions, and Stress Treatments

In this study, two genotypes of peanut with contrasting drought tolerance were selected, namely, drought-tolerant TAG 24 and drought-sensitive JL 24 genotypes. Seeds were obtained from ICAR – Directorate of Groundnut Research, Junagadh, Gujarat, India, and experiments were conducted at the Department of Biotechnology, Junagadh Agricultural University, Junagadh, India

The seeds were surface-sterilized with 0.1% HgCl2 for 1 min followed by washing with sterile water three times and placed on a moistened paper towel in sterile glass Petri plates for germination until radicles of 3–4 cm were visible (until 3 days). Each seedling was planted hydroponically on a perforated polystyrene sheet, placed over a plastic tray containing 4 L Hoagland’s nutrient solution,[61] and cultivated in a growth chamber with 16 h light (200 μM m–2 s–1, 26 °C) and 8 h darkness (24 °C) at 50% relative humidity for further 22 days. The nutrient solution was changed every third day, ensuring no depletion of oxygen and nutrients. Morphologically uniform 25 day old seedlings were selected for drought treatments. To create artificial drought stress conditions, a 20% PEG 6000 solution with the corresponding osmotic potential of about −0.49 MPa was used, according to Michel and Kaufmann.[62]

After the 25th day, seedlings were randomly divided into two groups (control and treatment). The control group continued to grow in Hoagland’s solution only, while another group was treated with Hoagland’s solution with 20% PEG 6000 for 24 h until visible wilting appeared. The upper-most peanut leaves and roots were collected (on the 26th day) from the 10 seedlings of all treated groups (2 genotypes × 2 tissues, i.e., leaf and root × 2 treatments, i.e., control and stress = 8 sample group) and pooled within sample groups. For root sampling, control and treated roots were washed with sterile distilled water to remove traces of chemicals and pat dry with paper towel before sampling. All of the samples were snap-frozen in liquid nitrogen immediately after collection and stored at −80 °C until metabolite extraction was performed.

Untargeted Metabolite Profiling

Extraction

Extraction of whole metabolites was performed as described earlier[63] with some modifications. One hundred milligrams of frozen ground samples (leaves/root) were homogenized in 1.5 mL of 100% methanol (HPLC grade, precooled at −20 °C) using a prechilled mortar–pestle. To that, 100 μL of ribitol (0.2 mg mL–1 stock in dH2O) was added as an internal quantitative standard and vortexed for 10 s. The mixture was shaken for 15 min at 70 °C in a water bath at 1000 rpm, followed by centrifugation at 11 000g for 10 min. The supernatant was collected in a glass vial, and 0.75 mL of chloroform (−20 °C) and 1.0 mL of distilled water (4 °C) were added and vortexed for 10 s. The mixture was centrifuged at 2200g for 15 min to separate polar and nonpolar phases. Upper (polar) and lower (nonpolar) fractions were collected in a separate test tube for drying under nitrogen stream in a turbo evaporator (Biotage, TurboVap LV).

Derivatization of Metabolites

Derivatization of the sample was performed as described earlier[64] with minor modifications. Briefly, to redissolve the dried extract, 50 μL of pyridine was added and sonicated for 10 min, followed by the addition of 100 μL of methoxyamine hydrochloride (20 mg mL–1 in pyridine) and gentle mixing; carbonyl components were protected by methoximation. The mixtures were further sonicated for 5 min and incubated with constant agitation at 37 °C for 90 min. For the tri-methyl-silylation (TMS) step, 250 μL of MSTFA (N-methyl-N-(trimethylsilyl)tri-fluoroacetamide) was added and the tube was sealed with paraffin and vortexed for 30 s. Mixtures were incubated at 37 °C for 1 h with constant agitation for derivatization. The derivatized extracts were kept at room temperature to cool down for at least 1 h before GC–MS analysis.

GC–MSAnalysis

The derivatized samples were analyzed by a GC–MS (gas chromatography–mass spectrometry, Shimadzu QP2010Plus, Japan) instrument connected to a mass selective detector (Shimadzu GC–MS-QP2010 SE, Japan) and operated according to the manufacturer’s instructions. The derivatized extract (1 μL) was injected into a column capillary (DB-17 MS, 30 m × 0.25 mm) using a splitless injection (230 °C, 1.5 min). The inlet temperature and ion source temperature were set at 280 °C and 230 °C, respectively. Helium gas (99.99% purity) was used at a flow rate of 1 mL min–1 as a carrier gas. The electron ionization of 70 eV was used in the full scan mode (50–1000 Da, m/z).

For chromatogram acquisition and peak deconvolution, GC–MS real-time analysis software (version Shimadzu) was used. Metabolites were putatively identified by similarity matching their mass spectra to spectra in the NIST 14 library (National Institute of Standards and Technology, Gaithersburg, MD)—a public database.[65] The preprocessing of total ion chromatograms (TICs), i.e., alignment, baseline correction, and integration, was carried out using ACD/Spec Manager v.12.00 (Advanced Chemistry Development, Inc., ACD/Labs, Toronto, Canada).[42] The data sets including sample information, retention time m/z, and peak intensity were formatted as CSV comma-delimited files and exported as an input file for MetaboAnalyst 4.0, data analysis software.

Targeted Metabolite Profiling

Extraction and Benzoylation

For the extraction of free polyamines from peanut leaf and root samples, a similar extraction method was performed for both samples. Samples (200 mg) were powdered in liquid nitrogen and extracted in 2 mL of chilled 5% HClO4 (perchloric acid).[66] The extract was incubated for 1 h in an ice bath followed by centrifugation at 15 000 rpm for 20 min. The supernatant containing the free polyamines was stored at −20 °C. These extracts remain stable for six months for polyamine analysis by TLC or HPLC.

Extraction was followed by the benzoylation step. To 250 μL of HClO4 extract, 1 mL of 2 N NaOH and 10 μL of benzoyl chloride were added. The mixture was vortexed for 10 s and incubated for 20 min at room temperature. Saturated NaCl (2.0 mL) was added and vortexed for 15–20 s. Standard polyamines (agmatine, putrescine, cadaverine, spermidine, and spermine) were also benzoylated similarly along with the extract. To the benzoylated extract, 2.0 mL of diethyl ether was added and centrifuged at 3000 rpm for 5 min. The ether phase (upper, 1.0 mL) containing benzoyl polyamines was collected and dried in a vacuum concentrator (Eppendorf Concentrator plus) without heating. The concentrated sample was redissolved in 100 μL of HPLC-grade absolute methanol. Standards were treated similarly, 70 ppm of each polyamine in the reaction mixture. The benzoylated samples were stored at −20 °C until the run was performed.

Estimation by HPLC

Polyamines were analyzed using HPLC (Waters 600 Controller), as explained by Flores et al.[67] The mobile phase consisted of methanol/water at a flow rate of 1 mL min–1, run isocratically at 60%. The benzoylated extracts were eluted through a reverse-phase (C18) column (4.6 × 250 mm, 5 μm particle size) at room temperature and detected at 254 nm (Waters Photodiode Array Detector 2996). The quantity of individual polyamines was calculated based on the area and concentration of the standard.

Extraction of Polyphenols

The fresh leaf and root samples (500 mg) were powdered in liquid nitrogen and extracted at room temperature for 20 min with 2 mL of 80% methanol solution (v/v).[68] The extract was filtered and evaporated to dryness under nitrogen stream in a turbo evaporator (Biotage, TurboVap LV). Immediately before analysis, the extracts were redissolved in 100 μL of absolute methanol.

Estimation of Polyphenols

Ultrahigh-performance liquid chromatography analysis was performed on an LC–MS/MS system (Waters Acquity UPLC-PDA, Milford). The Acquity UPLC system had a built-in autosampler and a binary solvent delivery system. The UPLC instrument was connected to a TQ mass spectrometer (Acquity). An MS system fitted with an electrospray ionization (ESI) source worked in negative ion mode and scan mode for multiple reaction monitoring (MRM). ESI ionization conditions are as follows: 1, source temperature 150 °C and desolvation temperature 350 °C; 2, negative ionization mode; 3, source voltage −3.2 kV (ESI– mode) and 3.00 kV (ESI+ mode). As curtain and auxiliary gas, high-purity nitrogen (>99.999%) was used. The chromatographic separation was performed on an Acquity UPLC BEH C18 column (2.1 mm × 100 mm, 1.7 μm particle size) at 35 °C. The sample injection volume was 10 μL, and separation was performed with a binary mobile phase at 0.4 mL min–1 during the analysis. The binary solvent system comprised 100% methanol and 1% (volume fraction) acetic acid in water as a gradient run (Table S4). These contents of polyphenols were determined by the UPLC–MS/MS method described by Zhang et al.[69]

For standard, 500 μg of standard phenolic compounds were dissolved in 1 mL of 80% methanol individually and then mixed. A total of 15 standards (cinnamic acid, caffeic acid, salicylic acid, gallic acid, ferulic acid, quercetin, catechol, chlorogenic acid, coumaric acid, syringic acid, kaempferol, vanillic acid, catechin, epicatechin, and epigallocatechin) were used in the present study.[42] A mixture of standards was made the same way as samples and identified based on its retention time and mass. The quantity of individual phenolics was calculated based on the area and concentration of standards.

Data Processing and Statistical Analysis

Data processing and statistical analysis were carried out by Metabo-Analyst 4.0 software (https://www.metaboanalyst.ca/),[70] an online statistical package. For statistical analyses, peak areas were taken into consideration. Data were normalized to the internal standards (ribitol); data transformation was none; and Pareto scaling was used to put all variables on equal footing, minimize variable redundancy, and adjust for measurement errors.[71] Pareto scaling helps to increase the amplification of low-abundance ions without amplification of raw data noise. Heat maps were generated based on the Pearson distance measure and the Ward clustering algorithm; the top 25 metabolites were shown for control versus drought treatments to visualize relative levels. To visualize important metabolites among the performed groups, multivariate tests, viz., partial least-squares discriminant analysis (PLS-DA) and principal component analysis (PCA), were employed. PLS-DA is a supervised method used to analyze large data sets. The variable importance in projection (VIP) score ranks the overall contribution of each variable using a significance level of p ≤ 0.05. Dendrogram analysis was performed to reveal the relationships of metabolites. The important metabolites were identified by using PLS-DA, VIP scores, and heat maps.[22,72,73] For polyamine and polyphenol analysis, the concentration was taken into consideration. Data were normalized by Pareto scaling, and heat maps were generated to identify relative levels in control and stress treatments in both genotypes.

Pathway Analysis

Pathway analysis was carried out in the same statistical package, MetaboAnalyst 4.0. Putatively identified untargeted metabolites were imported in one column compound list and proceeded as per the guidelines provided. The Arabidopsis thaliana pathway library was used for pathway analysis. Distribution and p-values < 1 represented notable enrichment of certain metabolites in a pathway. Many pathways were tested at the same time, so the statistical p-values from enrichment analysis were adjusted through the false discovery rate (FDR) estimation.