Metabolomic analysis of date palm seedlings exposed to salinity and silicon treatments

ABSTRACT

Silicon is known to promote plant growth as well as stress tolerance of plants. The current study was undertaken to assess the growth promoting effect of silicon on date palm seedling development as well as its ability to abate some of the negative effects of salinity. In this study, date palm seedlings were treated with silicon and sodium chloride salts, and the effect of these salts on some physiological parameters of the plants was determined. In addition, a global nontargeted metabolomics analysis was performed for the leaf and root tissues using liquid chromatography-mass spectrometry (LC-MS). The results showed that under non-stress conditions, silicon treatment enhanced the growth of the date palm seedlings, however, under salinity, silicon slightly mitigates the negative effects of salt stress on the date palm seedlings although it enhances the potassium accumulation under this condition. The global metabolomics analysis has identified a total of 1,101 significant differentially accumulated (p, q ≤ 0.05) metabolites in leaves and roots under silicon, salinity or their combination. A differential pairwise metabolic profile comparison revealed the accumulation of distinct metabolites in response to silicon and salinity treatments such as antioxidant compounds pyridoxine, cepharanthine, allithiamine, myristic acid and boldine; osmoregulators such as mucic acid; along with the accumulation of detoxification intermediates such as S-D-lactoylglutathione, beta-cyano-L-alanine and gamma-glutamyl-conjugates. In addition, histochemical analyses revealed that application of silicon significantly (p ≤ 0.05) enhanced the formation of the Casparian strip. Identification of the differentially accumulated metabolites could offer an insight into how silicon is able to promote growth and salinity tolerance in date palms.

KEYWORDS:

• Salinity
• date palm
• metabolomics
• antioxidants
• detoxification
• tolerance
• silicon

1. Introduction

Salinization is estimated to affect more than 20% of total cultivated land and 33% of irrigated land worldwide, and this trend is increasing at a rate of 10% annually, which may result in a 30% land loss over the next 25 years, and up to 50% by the year 2050.¹ Coincidentally, some of the date palm cultivation areas are in heavily salt-affected areas.²

Silicon is the second most abundant element in the earth’s crust, constituting about 28% of the total elements present,³ in the form of silicate minerals and various forms of silicon dioxide.⁴ Plants take up silicon from the soil in the form of uncharged bioavailable monomeric H₄SiO₄ or its ionized form Si(OH)₃O⁻, which mainly occurs when the pH value is more than 9.⁵ The concentration of silicon in plant tissues varies from one species to another. For example, Oryza sativa and Hordeum vulgare accumulate more silicon than Helianthus annuus, Benincasa hispida and Lycopersicon esculentum^6,7 and so far silicon toxicity has not been reported.⁸ The soil where date palms grow is mainly composed of quartz sand, which is 80% silica, a plant bio-inactive form that is not readily available for plant uptake.

Despite the fact that silicon is not an essential element required for plant growth and development, for the last two centuries the use of silicon fertilizers has been reported to play an important role in agriculture.⁹ Silicon ions are thought to enhance several beneficial traits in plants grown under normal conditions or abiotic stress conditions such as drought, salinity, cold, heat, heavy metals, or even under biotic stress caused by insect pests and other pathogens.^10,11 For instance, in tomato, silicon promoted flowering and fruit set,¹² enhanced salt and drought tolerance and provided protection against disease-causing pathogens.¹³

Generally, silicon is assumed to help plants via the enhancement of the structure of their organs through the formation of polymerized silicates in the plant tissues¹⁴ and via the modulation of gene expression.¹⁰ In addition, silicates are able to mitigate the effects of salinity by increasing the antioxidant enzymatic activity.¹⁵ In some plant species, silicates may enhance the formation of Casparian bands in the epidermis and endodermis, which act as a barrier preventing the non-selective accumulation of salts into the stele via the apoplastic pathway.¹⁶

In date palm, metabolite profiles were previously carried out to identify early signs of heat and drought,¹⁷ and also to assess the quality and the nutrient value of the fruits.^18-20 The present study aimed at analyzing the beneficial effects of silicon in date palm under control conditions, salt stress or their combination, and to identify the associated metabolites in the respective treatments. The date palm seedlings were exposed to NaCl, silicon and their combination, and changes in physiological and histological parameters such as photosynthesis, leaf area, dry weight, sodium, and potassium concentration and Casparian strip formation were investigated. Additionally, global untargeted metabolite profiles were analyzed in order to gain an insight into the metabolic basis of salt-tolerance promoting role of silicon in date palm. The results showed that silicon triggered the production of antioxidant compounds such as boldine, allithiamine, cepharanthine, myristic acid and pyridoxine along with the accumulation of detoxification intermediates such as S-D-lactoylglutathione, beta-cyano-L-alanine, and gamma-glutamyl-conjugates. Previous studies have indicated the accumulation of few metabolites in response to salinity treatment in date palm,^21,22 however, this is the first report that globally analyzed metabolites in the leaf and root tissues of this plant when exposed to salinity and silicon treatments.

2. Materials and methods

2.1 Plant growth conditions and silicon treatment

Date palm (Phoenix dactylifera L., cultivar Khalas) seeds were surface-sterilized with 1% commercial bleach for 5 min, followed by 70% ethanol for 10 min, after which they were rinsed and germinated on moist sterile vermiculite in a growth chamber at 37°C for about 10 days under dark, as previously described.²³ Once the radicle had emerged, the seedlings were transferred to pots containing potting compost, which were kept in a growth chamber and maintained at 32°C/30°C with a photoperiod of 16/8 h. The seedlings were divided into four sets of treatments; a control treatment in which the plants were irrigated with tap water having an electrical conductivity of 50 mS/m, the second set received 300 mM NaCl (salt stress), the third set received 5 mM Na₂SiO₃ (+Si) alone (silicon treatment) and the fourth set received both 300 mM NaCl as well as 5 mM sodium silicate (Na₂SiO₃) (NaCl+Si combined treatment). The NaCl and silicon treatments were administered along with the irrigation on a weekly basis for 8 weeks.

2.2 Photosynthesis-associated parameters and biomass accumulation

The net photosynthesis (A), internal CO₂ concentration (ci), stomatal conductance (gs) and transpiration (E) were measured using the LCpro-SD advanced photosynthesis measuring system (ADC Bioscientific Ltd., UK). The leaf surface area was measured using the LI-COR Biosciences LI-3000A Leaf Area Meter (LI-COR Biosciences, Nebraska, USA). For the leaf and root dry mass accumulation, the plants were harvested, rinsed under running tap water to remove any soil particles and the roots and leaves were separated, dried at 70°C for 24 h in an oven, and the dry weight was recorded.

2.3 Sodium and potassium content in the plant tissues

The plant samples were cut into smaller pieces and placed in 25 ml glass vials to which 10 ml of 1M HCl was added. The vials were maintained at 200°C using a hotplate for 4 h. The samples were then filtered into volumetric flasks using Whatman 42 filter papers. The Na⁺ and K⁺ concentrations were measured following the previously described protocol,²⁴ using an SYSTRONICS Flame Photometer 128 (SYSTRONICS Ltd., Gujarat, India).

2.4 Histochemical root staining

For histochemical staining assay, date palm seedlings were hydroponically grown in a solution containing the commercial liquid fertilizer 20-05-10 NPK, along with the trace elements boron, zinc, iron, manganese, and copper, as the nutrient source, together with the treatments of NaCl (50 mM) and silicon (5 mM). For the histochemical staining of the roots, a distance of 10 cm was measured from the root tip and hands-free sections were made in this root zone using a sharp blade and stained as described previously.²⁵ The sectioned roots were stained using 0.1% (w/v) berberine chloride for 60 min., after which the excess stain was removed. Subsequently, the tissue was counterstained with 0.5% (w/v) aniline blue for 30 min. The stained root sections were then mounted in 0.1% (w/v) FeCl₃ in 50% (v/v) glycerin and viewed under a fluorescent microscope (Nikon Eclipse Ni, Nikon, Tokyo, Japan) with UV illumination and excitation filter G 365, chromatic beam splitter FT 395 and barrier filter LP 420. For the comparisons of the formation of Casparian bands, three readings were taken around the endodermis from each root section and each treatment was replicated five times.

2.5 Global metabolite analysis

Samples were divided into eight different groups based on the treatments and the tissue types. The tissues of the control plants were then separated into leaves (CL) and roots (CR). The salinity treated plants were then divided into leaves (NL) and roots (NR). The Silicon treated plants were then divided into leaves (SiL) and roots (SiR) The plants in which both the salt- and silicon-treatment were further divided into leaves (NSiL) and roots (NSiR). Each group composed of three biological replicates. Each replicate included tissues from three different plants. At the end of the treatment (120-day old seedlings), the leaves and roots were harvested, flash-frozen in liquid nitrogen and freeze-dried for 48 h and used for the metabolite analysis.

The untargeted metabolite and primary data analyses were carried out at the Proteomics and Mass Spectrometry Facility, Donald Danforth Plant Science Center, Saint Louis, USA, as a service provider. Leaf and root tissues were separately ground and the metabolites were extracted in 80% methanol by vortexing for 10 min, centrifuged, and the aqueous phase was filtered using an 0.8-μm spin filter made of polyethersulfone (PES), after which they were dissolved in 10 mM NH₄HCO₃ in water and 10 mM NH₄HCO₃ in 95% acetonitrile, as solvents. One microliter of each sample was loaded and then analyzed using liquid chromatography-mass spectrometry (LC-MS), via hydrophilic interaction liquid chromatography (HILIC) on a 0.5 x 150-mm ZIC-pHILIC column and reverse-phase liquid chromatography (RPLC) on a 0.5 x 100-mm PLRP-S column. The complete system setting used in the HILIC and RPLC-LC-MS, and library searching were listed in Table S1. Briefly, LC-MS/MS was carried out on a high-resolution accurate Thermo Q-Exactive Orbitrap mass instrument (ThermoFisher Scientific, Waltham, MA). A gradient chromatographic separation and positive and negative polarity-switching ion mode were used in this analysis. Since no internal standards were used, the data were normalized using quantile normalization. Raw data files were converted to mz5 format using ProteoWizard version 3.0.9987.²⁶ Feature finding (a/k/a peak picking) was performed using Elements version 1.4.2.²⁷

The results from the LC-MS analysis were analyzed using Scaffold Elements viewer software version 2.1.1.²⁶ The log₁₀ value of the average precursor intensity (area under the peak) of each analyte was used as the concentration value for each analyte. The metabolite groups and the metabolite clusters were filtered based on the confidence in the analyte identity value (ID score) with a threshold value of ≥ 0.7. The various metabolites within the groups were selected based either on the highest ID score or on the potential biological function in plants. Significant differences in the accumulation of metabolites among the different samples were calculated using a t-test (p, q ≤ 0.05). The significance level of multiple testing was corrected using the false discovery rate (FDR) following using the Benjamini-Hochberg procedure at a level of q ≤ 0.05.²⁸ These metabolites were then identified via the linked ChemSpider database.²⁹ After identification, the metabolites were functionally characterized by searching for the corresponding metabolite and metabolic pathways in the Human Metabolome Database (HMDB).³⁰ Mapping of the metabolites onto the corresponding pathways was carried out using the Kyoto Encyclopedia of Genes and Genomes (KEGG) database,³¹ using the default settings. The accumulation of differential amounts of metabolites in different samples was illustrated by a heat map using PermutMatrix version 1.9.3. software,³² using the log₁₀ value of the area under the peak of each analyte, the Euclidean distance dissimilarity distance, and the multiple fragment heuristic seriation rule. A principal component analysis (PCA) was carried out in order to reveal the statistical correlations between metabolites produced in the date palm leaf and root tissues grown under different treatments, using Past 3 software.³³ The log₁₀ values of the area under the peak of each analyte and the default setting of the software was used in this analysis. The presence of a specific analyte within the metabolomic profile of the samples was demonstrated by point, illustration, explanation (PIE) analysis using Venn diagrams via Venny 2.1 software.³⁴ This analysis will only show the presence or absence of specific analyte in each of the treatments, but not the differentially accumulated metabolites in the plant tissues due to the different treatments.

2.6 Statistical analysis

The results of the growth parameters under different treatments were compared and the significance verified using a one-way ANOVA t-test and using univariant analysis and Tukey’s post hoc test, in SPSS software version 21.

3. Results

3.1 Silicon enhanced plant growth under salinity

Salinity treatment at a concentration of 300 mM NaCl was the most effective salinity stress for date palm seedlings, while 5 mM Na₂SiO₃ was the most effective concentration on plant growth. The growth and the biomass of the date palm seedlings were greatly affected in plants exposed to salinity stress and treated with silicon (Figure 1(a)).

Figure 1. The effect of silicon and salinity on the growth of date palm seedlings (a). Growth changes in date palm seedlings grown under silicon (Si), control, silicon and saline (NaCl+Si) and saline (NaCl) conditions. These changes in the leaf area (b), leaf fresh weight (c) and root dry weight (d) of date palm seedlings subjected to silicon, salinity, and silicon combined with salinity treatments. The bars represent the means (n = 9 ± SE) and significant differences at p < .05 are marked by asterisks.

The leaf area was significantly (p < .05) increased in plants treated with silicon (+Si) alone as well as in plants treated with silicon and salinity (NaCl+Si) (Figure 1(b)). The dry mass accumulation of the leaf and root tissues was negatively affected by salinity; however, it was mildly increased in leaves treated with either silicon (+Si) or silicon combined with salt stress (Figure 1(c)). The dry mass accumulation of the root tissue was also significantly (p < .05) increased in plants treated with silicon alone (+Si) (Figure 1(d)).

The effect of silicon treatment on the photosynthetic parameters such as the net photosynthesis

(A), internal CO₂ concentration (ci), stomatal conductance (gs) and transpiration (E) was measured in leaves of plants treated with silicon with or without salt stress. Interestingly, A (p < .05) and ci (p < .001) were significantly increased in plants treated with silicon alone compared with the control conditions (Figure 2(a,b)). Treatment with salinity led to a slight increase in gs in silicon-treated plants. However, no significant change was observed in plants exposed to saline conditions (Figure 2(c)). The results also showed that there was a significant (p < .01) increase in E when plants were treated with silicon alone but did not show a significant effect on the plants grown under saline conditions (Figure 2(d)).

Figure 2. Changes in photosynthesis-associated parameters due to salinity and silicon treatments. The parameters include the net photosynthesis (A) (a), internal leaf CO₂ concentration (ci) (b), stomatal conductance (gs) (c) and transpiration (E) (d). The values represent the means (n = 9 ± SE) and significant differences at p < .05, p < .01, p < .001 are marked by asterisks as (*), (**), and (***), respectively.

3.2 Silicon treatment enhanced K⁺ uptake in roots

The Na⁺/K⁺ homeostasis plays an important role in salinity tolerance therefore; the concentration of these ions was measured in the leaf and root tissues exposed to different treatments. The Na⁺ levels in leaves and roots were significantly (p < .05) increased under salt stress. The Na⁺ levels were slightly decreased in leaves of plants exposed to silicon alone (Figure 3(a)). However, the Na accumulation pattern did not differ between salinity and salinity plus silicon-treated group. In general, salinity led to a significant (p < .05) negative effect on K⁺ accumulation in the plants. However, the combined salt and silicon treatments resulted in a significant (p < .05) increase in K⁺ concentration in root tissue (Figure 3(d)) but a mild decrease in leaf tissue (Figure 3(c)).

Figure 3. Changes in Na⁺ and K⁺ concentrations in response to salt stress and silicon treatments in leaves (a) and roots (b). The values represent the means (n = 9 ± SE) and significant differences at p < .05, and p < .001 are marked by asterisks as (*), and (***), respectively.

3.3 Silicon enhanced the formation of casparian strips in root tissues

The Casparian strip plays a crucial role in the selective apoplastic bypass of salts under salinity. To assess whether the formation of Casparian strips in roots was altered under salt stress or salt stress combined with silicon treatment, histochemical staining was performed. The thickness of the Casparian strips was significantly (p < .05) increased in plants treated with silicon compared to the untreated control plants (Figure 4). Similarly, it was also significantly increased in plants exposed to salt stress combined with silicon treatment compared with salt stress-treated plants (Figure 4).

Figure 4. The changes in the width of the Casparian strip observed under the fluorescent microscope in roots of plants grown under different treatments. The stained root images were taken from control plants (a), salinity-treated plants (b) and silicon alone treated plants (c), and salt stress plus silicon-treated plants (d). The measurements were statistically analyzed for different samples (e). Changes in the width were measured in μm. The values represent the means (n = 15 ± SE) and significant differences at p < .001 are marked by asterisks.

3.4 Metabolic profiles of date palm

In general, the silicon treatment has decreased the negative effects of salinity on date palm seedlings (Figures 1–4). To gain an insight into the metabolic basis of this observation, global untargeted metabolite profiles of leaf and root tissues from the plants exposed to different treatments (control, silicon, NaCl and NaCl+silicon) were obtained. The analysis revealed a total of 1,043 significantly (p, q ≤ 0.05) accumulated metabolites. Of these, 544 and 499 metabolites were obtained using the RPLC and HILIC techniques, respectively. Only metabolites with FDR corrected p (q) ≤0.05 were included in the analysis. This analysis revealed the presence of a high number of metabolites in date palm tissues, however, the current study mainly focused on metabolites that were previously identified from other plant species.

A principal component analysis (PCA) which explained about 64% of the variance, suggested a statistical correlation between replicates within each treatment group as well as between different treatment groups. The replicates within each metabolic group were clustered together on the PCA chart plotter, suggesting a high level of homogeneity within the metabolic profiles of the replicates of each group (Figure 5).

Figure 5. Principal component analysis (PCA) illustrating distances between metabolites identified from a group of three replicate samples of leaves (CL) and roots (CR) when the plants were irrigated with distilled water, leaves (NL) and roots (NR) when the plants were irrigated with 300 mM NaCl solution, leaves (SiL) and roots (SiR) when the plants were irrigated with a silicon solution only and leaves (NSiL) and roots (NSiR) when the plants were irrigated with a salt solution in the presence of silicon. The first two coordinates explained about 64% of the variance.

Additionally, the treatment groups were distributed on the chart plotter based on the tissue type and the treatment. While metabolic profiles identified from leaf tissues (CL, SiL, NL, and NSiL) were localized on the left-hand side of the chart, metabolic profiles from the root tissues (CR, SiR, NR, and NSiR) were localized on the right-hand side of the chart. Similarly, while the profile groups identified from control group (irrigated with water) or a silicon solution (CL, SiL, CR and SiR) were localized on the upper part of the chart, whereas most of the profiles identified from salinity-treated plants (NL, NSiL, NR and NSiR) were localized on the lower parts of the chart plotter. Interestingly, the metabolic profiles of the roots of silicon-treated (SiR) and untreated (CR) plants were clearly separated on the chart plotter. Similarly, a lesser extent separation was evident for the roots exposed to salt stress or combined salt stress and silicon treatment. However, the separation between silicon and without silicon treatments was not clear for the leaf tissues, i.e., between control and silicon-treated plants or between salt stress and combined salt stress and silicon-treated plants (Figure 5).

3.5 Metabolites accumulating in the plants in response to salinity and silicon treatments

A total of 1,101 differentially accumulated metabolites were identified from leaf and root tissues in this study. Of these, 836 and 839 were identified from the leaves and roots, respectively. Among them, 345 metabolites were shared between both tissue types under different conditions tested (Supplement Table S2). Of the 836 unique metabolites identified in the leaf tissue subjected to four different treatments (control, silicon, salinity, and combination of salinity and silicon), 651 metabolites were shared between all treatments (Figure 6). Similarly, 578 metabolites in roots were shared between all four conditions tested. When compared with the plants grown under control conditions, only nineteen and six metabolites were exclusively accumulated in the leaf and the roots respectively, in the silicon-treated samples. Likewise, upon the comparison between salt stress alone and salt stress combined with silicon treatment, 12 and 17 metabolites were exclusively accumulated in the leaves and roots respectively, in the plants grown under salinity combined with silicon treatment (Supplementary Table S3 and Supplementary Table S4). Interestingly, there were 47 metabolites exclusively accumulated in the leaves of plants under salinity, of which six were only detected in NaCl-treated leaves and twelve were only detected in salt stress combined with the silicon-treated leaves. However, there were 29 metabolites in the leaves common between both salt stress conditions, i.e., presence or absence of silicon (Figure 6).

Figure 6. The distribution of a number of metabolites identified from leaves (a) and roots (b) of the plants grown under control, salinity, silicon or salinity combined with silicon treatment. The number of metabolites produced in leaves (CL) or roots (CR) of control plants, leaves (NL) or roots (NR) of plants treated with salinity, leaves (SiL) or roots (SiR) of plants treated with silicon alone or leaves (NSiL) or roots (NSiR) of plants treated with silicon and salinity, are illustrated in the Venn diagram.

The abundances of the metabolites identified in this study were varied qualitatively and quantitatively between the tissues during different treatments. While some of them were simply present or absent (Figure 7), others were present at different levels in the samples (Supplementary Figure S1 and Supplementary Table S5).

Figure 7. Heat map depicting the differential accumulation of selected metabolites and their classification in the leaves or roots of plants grown under control, salinity, silicon or salinity combined with silicon treatment. The heat map in leaves (CL) or roots (CR) of control plants, leaves (NL) or roots (NR) of plants treated with salinity, leaves (SiL) or roots (SiR) of plants treated with silicon alone or leaves (NSiL) or roots (NSiR) of plants treated with silicon and salinity. The heat map was constructed using the log₁₀ of the mean value of three replicates of the precursor intensity (area under the peak) of three replicates of each analyte.

3.6 Differentially accumulated metabolites in leaves in response to salinity

Pairwise analysis of the metabolic profiles revealed 133 differentially accumulated metabolites in the leaves of plants exposed to salt stress (Supplementary Table S6). Some of these metabolites may play important biological roles under salinity. For example, succinylacetone, a cytotoxic metabolite of tyrosine (Supplementary Figure S1), the nitric oxide inhibitor symmetric dimethylarginine (Figure 8), were slightly enhanced in the leaf tissues in response to salinity treatment. Additionally, the various dipeptides including phenylalanyl-methionine, glutamyl-phenylalanine, glutamyl-tyrosine, and aspartyl-glutamine, were repressed in the leaves under salinity conditions (Figure 8). The tryptophan metabolism intermediate 2-aminomuconic acid, the GMP synthase inhibitor decoyinine and the salicylic acid beta-D-glucoside, which is a storage form of salicylic acid were some other differentially accumulated metabolites (Supplementary Table S6). In addition, there were 47 metabolites that were specifically accumulated in the leaves under salinity. These include metabolites such as phosphatidylethanolamine lyso 18:2, along with the polyether compound polypropylene glycol (Supplementary Table S3).

Figure 8. Boxplot presentation of some differentially accumulated metabolites involves oxidation scavenging pathways in date palm. The x-axis on the graphs represent the treatments while the y-axis represents the log₁₀ value of the area under the MS peaks.

3.7 Differentially accumulated metabolites in roots in response to salinity

Differential pairwise analysis of the metabolic profiles showed that there were 187 significantly (p, q ≤ 0.05) accumulated metabolites in roots in response to salinity. Some of them have known function in salinity tolerance in plants. For example, dipeptide and tripeptide metabolites ala-gly-ala and alanyl-glutamine, N-a-acetylcitrulline, an intermediate metabolite of the arginine biosynthesis,³⁵ phenolamide 4-coumaroylputrescine, hypochlorous acid inhibitor 4-aminobenzoic acid hydrazide,³⁶ 4-methylumbelliferone which is a coumarin derivative which inhibits lateral root formation³⁷ and sakuranetin, which is a jasmonic acid-induced phytoalexin which enhances stress tolerance³⁸ (Supplementary Table S7). A total of 51 metabolites were specifically accumulated in roots due to salinity (Figure 6, Supplementary Table S4). Among these metabolites, the abscisic acid (ABA)-oxidized metabolite 7′-hydroxyabscisic acid, the GMP synthase inhibitor decoyinine and the antioxidant dihydrolipoate, which is a reduced form of lipoic acid are some of the worth mentioning metabolites.

3.8 Differentially accumulated metabolites in leaves in response to silicon treatment alone

The pairwise analysis was carried out in order to investigate the role of silicon in plant growth enhancement. The analysis revealed the presence of 111 differentially accumulated metabolites with significant (p, q ≤ 0.05) quantities in leaf tissues of the plants grown under control conditions (Supplementary Table S8). These metabolites included the gamma-glutamyl conjugate L-gamma-glutamyl-L-leucine,³⁹ the reactive oxygen species (ROS) quencher pyridoxine, also known as vitamin B6,⁴⁰ the dipeptide glu-leu, asparaginyl-arginine and phenylalanyl-arginine, the antioxidant alkaloid cepharanthine⁴¹ and the antimicrobial metabolites glepidotin C⁴² and paromomycin.⁴³ The accumulation of these antimicrobial metabolites was affected due to silicon and salinity treatments (Supplementary Figure S1). Importantly, 19 metabolites were exclusively accumulated in silicon-treated leaves and these compounds were absent in leaves from control samples (Figure 6). These metabolites included the antifungal aromatic anofinic acid,⁴⁴ the plant hormone salicylic acid and a jasmonic acid derivative, epi-4ʹ-hydroxyjasmonic acid, which was shown to be involved in mediating plants’ defense against pathogens and feeding pests.⁴⁵ The list also included the flavonoid ketohexosefuranoside pinnatifinoside-A and the phenolic antioxidant epicatechin 3-glucoside (Supplementary Table S4).

3.9 Differentially accumulated metabolites in roots in response to silicon treatment alone

In comparison with the silicon untreated roots, a total of 152 differentially accumulated metabolites with significant (p, q ≤ 0.05) amounts were identified in root tissues, when date palm seedlings were treated with silicon (Supplementary Table S9). These metabolites included the plant hormone salicylic acid, the salt-tolerance inducing amino acid L-asparagine, the antimicrobial and antioxidant flavonoid bolusanthin, the intermediate detoxified metabolite of formaldehyde S-formylglutathione, the gamma-glutamyl conjugate gamma-glutamyltyrosine and the dipeptides aspartyl-glutamine and arginyl-asparagine,⁴⁶ the cytoprotective flavonoid glycoside nicotiflorin,⁴⁷ antioxidant metabolites such as the flavonoid epicatechin,⁴⁸ the vitamin B1 allithiamine,⁴⁹ and the alkaloid boldine.⁵⁰ These metabolites showed a differential pattern of accumulation when the plants were treated with silicon under control conditions (Supplementary Figure S1). Additionally, there were six analytes that were specifically accumulated in the roots, such as the anticancer agent glucoraphanin⁵¹ and the antimicrobial and antioxidant isocoumarin coriandrone-C⁵² (Supplementary Table S4).

3.10 Differentially accumulated metabolites in leaves exposed to salinity combined with silicon treatments

Differential metabolic profile analysis revealed the presence of 58 significantly (p, q ≤ 0.05) accumulated metabolites in leaf tissues in response to salinity and silicon treatments. These included metabolites with potential biological functions in plants, such as the amino acid L-histidine which induces pathogen resistance in Arabidopsis,⁵³ along with its metabolism intermediaries imidazole lactic acid⁵⁴ and 4-imidazolone-5-propionic acid,⁵⁵ the potential chelate molybdenum 2-oxoarginine,⁵⁶ which is an intermediate metabolite derived from the metabolism of arginine, the antioxidant alkaloid cepharanthine,⁵⁷ terephthalic acid,⁵⁸ which is the precursor for benzoate biosynthesis and the oxidative stress inducer phthalic acid⁵⁹ (Supplementary Table S10). In addition, 12 metabolites were specifically accumulated in leaves in response to silicon and salinity. These included the antimicrobial metabolite rifapentine⁶⁰ and antioxidant diterpene glycosides in the form of hovenidulcioside B1⁶¹ (Figure 6, Supplementary Table S3).

3.11 Differentially accumulated metabolites in roots exposed to salinity combined with silicon treatments

The differential metabolite analysis revealed the presence of 190 significantly (p < .05) accumulated metabolites in roots when plants were treated with silicon and grown under salinity (Supplementary Table S11). The list included saturated myristic fatty acid, mucic acid, and glucaric acid, which are hexaric acids formed by the oxidation of sugars,⁶² the two methylated derivatives of myo-inositol, (-)-bornesitoland the soluble sugar L-pinitol. These analytes were slightly induced in response to silicon treatment in the plants (Supplementary Figure S1). In addition, hydrogen cyanide detoxification intermediates S-D-lactoylglutathione,⁶³ beta-cyano-L-alanine and L-3-cyanoalanine, which was induced in the leaf tissues due to silicon treatment under control and salinity treatments (Figure 8). The diketogulonic acid, which is a hydrolyzed product of dehydroascorbic acid, isocitric acid and citrate, which are metabolites of the tricarboxylic acid cycle (TCA),⁶⁴ the antimicrobial phytoalexin glyceollidin I,⁶⁵ and selenium sulfide, which may be a byproduct of the oxidation of glutathione⁶⁶ were significantly accumulated (Supplementary Table S11).

There were 15 metabolites specifically detected in silicon-treated roots under salinity (Figure 6, Supplementary Table S5). These metabolites included the condensed tannin cinnamtannin A2, with potential antimicrobial activity,⁶⁷ mucic acid, the ascorbic acid metabolite 2,3-diketo-L-gulonate and the freezing-induced glycoglycerolipid monogalactosyldiacylglycerol (MGDG 36:5).⁶⁸

3.12 Detoxification pathways enriched in salinity combined with silicon treatment

Metabolic pathway mapping of metabolites accumulated in leaf and root tissues of seedlings treated with silicon suggested that some of them could play a role in three main detoxification pathways. The detoxification pathways included the degradation of the cytotoxic dicarbonyl methylglyoxal (MG), which is a byproduct of various metabolic processes through the glyoxalase I, II and III of the glyoxalase systems. This resulted in the conversion of the glutathione-bound MG to S-D-lactoylglutathione, with the help of lactoylglutathione lyase.⁶³ S-D-lactoylglutathione is further broken down through glyoxalase II with the help of hydroxyacylglutathione hydrolase enzyme, to D-lactate (Figure 9). Indeed, our results showed an increase of the S-D-lactoylglutathione in the silicon-treated plant tissues when grown under control and salinity conditions.

Figure 9. An illustration showing the involvement of metabolites isolated from silicon-treated plants grown under salinity in the methylglyoxal (MG) detoxification mechanism, via the glutathione (GSH)-dependent glyoxalase I and glyoxalase II and the glutathione (GSH)-independent glyoxalase III pathways.

The second potential detoxification metabolic pathway could be through the gamma-glutamyl cycle and via gamma-glutamyl-cotransferase (GGCT), along with the 5-oxo-prolinase pathway.⁴⁶ These two pathways are involved in glutathione catabolism and amino acid recycling processes in the cell. The various gamma-glutamyl-conjugated compounds and 5-oxo-prolines found in the current study may be involved in these pathways, it was noted that there was an increase in seven out of nine gamma-glutamyl-conjugates in the leaf tissues and five out of seven in the roots under salt stress condition. There was a further increase in five and four gamma-glutamyl-conjugates in the leaf and roots respectively when silicon was supplemented under salt stress condition (Supplementary Table S12). In the detoxification and recycling pathway, the extracellular apoplastic and vacuolar oxidized-glutathione and glutathione-bound toxic metabolites are detoxified and recycled with the help of gamma-transferase (GGT), gamma-glutamyl-cotransferase (GGCT) and the 5-oxo-prolinase enzymes (Figure 10). These observations suggest that silicon enhances the detoxification pathway of toxic metabolites.

Figure 10. An illustration showing the involvement of metabolites isolated from silicon-treated plants grown under salinity in glutathione degradation and recycling, via the extracellular and vacuolar gamma-glutamyl cycle (a) and cytosolic GGC and 5-Opase (b) pathways.

The third potential pathway could be the detoxification of hydrogen cyanide in plants. In this pathway, the nitrile detoxification intermediates are produced via the cyanide detoxification pathway, which involves beta-cyano-L-alanine.⁶⁹ This metabolite was accumulated in silicon-treated leaves in control conditions, with further enhancement under salt stress treatment. On the other hand, the accumulation of beta-cyano-L-alanine was reduced in the roots under both control and salinity conditions on silicon treatment (Figure 8). This may suggest that the production of hydrogen cyanide, a by-product of the ethylene biosynthesis in the salt-treated plants was inhibited by silicon treatment. Additionally, the various metabolites identified in this study were mapped on the cyanoamino acid metabolism pathway (Figure S2).

4. Discussion

Date palm is a slow-growing and environmentally stimulating plant. However, silicon has an enhancing effect on the growth of the plant, including increasing the leaf area and the dry mass accumulation of the leaves and roots of plants, when grown under control and saline conditions. This increase could be attributed to the silicon-mediated ameliorating effect of the salt stress in date palm.

Although the accumulation of Na⁺ in leaves and roots was not significantly affected by silicon treatment under control or saline conditions, there was a significant (p, q ≤ 0.05) increase in the concentration of K⁺ which resulted in better/higher K⁺/Na⁺ ratio, which is one of the important traits contributing to the salt tolerance.⁷⁰

This increase in K⁺ accumulation in roots under salt stress could be due to the presence of silicon, which could enhance Casparian band formation in the roots. This type of enhancement was previously observed in other plant species⁷¹ and suggests the direct involvement of silicon in the salinity tolerance of date palm.

Metabolomics analysis was carried out to better understand the metabolic basis of silicon-mediated growth enhancement as well as salt tolerance in date palm. This metabolomic analysis revealed that treatment with silicon under salinity leads to the accumulation of some antioxidant metabolites in the leaves of date palm seedlings. For example, pyridoxine, which is also known as vitamin B6, is an important antioxidant cofactor that efficiently quenches ROS under stress conditions.^72,73 Similarly, the alkaloid cepharanthine was shown to have antioxidant and radical scavenging activity.⁵⁷ Alike, the alkaloid boldine has previously been reported to have a high reactivity toward free radicals and antioxidant properties.⁷⁴

The metabolites that were significantly accumulated in the roots of silicon-treated plants under salt stress conditions included myristic acid, which controls various redox imbalances originating from stress conditions in plants via N-myristylation, including the myristylation of SOS3 proteins, an important sodium detoxification pathway in plants.⁷⁵ In addition, ascorbate, a water-soluble form of vitamin C, which is a potent antioxidant that quenches ROS, was observed to be accumulated. The oxidization of ascorbate leads to the production of a reversible form of ascorbate free radical, which can be further oxidized to form dehydro-l-ascorbic acid, and further hydrolysis of dehydro-l-ascorbic acid produces 2,3-diketo-L-gulonate. All three forms have been shown to have antioxidant properties.⁷⁶ It was also suggested that the catabolism of ascorbic acid correlates with the rate of cell expansion,⁷⁷ and that the metabolites of 2,3-diketo-L-gulonate play a role in pathogen defense.⁷⁸

Date palm seedlings treated with silicon and grown under control conditions also accumulated plant biotic defense metabolites such as the antifungal aromatic anofinic acid,⁷⁹ the broad-spectrum antifungal agent glucoraphanin⁸⁰ and the jasmonic acid derivative epi-4ʹ-hydroxyjasmonic acid.⁸¹ Previous reports showed that silicon enhances tolerance toward pathogens and pests.⁸² In addition, the antimicrobial rifapentine⁸³ was highly accumulated in roots of plants treated with silicon under saline conditions. The upregulation of various metabolites that are associated with biotic stress tolerance in silicon-treated seedlings may also have a role in tolerating abiotic stresses such as salinity. Salicylic acid was also accumulated in the roots due to silicon treatment under control conditions. The accumulation of this metabolite, which has previously exhibited a positive effect on the root growth of other plant species,⁸⁴ is consistent with the phenotypic observations obtained in this study, where silicon-treated plants exhibited greater root biomass compared with the control plants (Figure 1(d)).

The metabolomic profiles in response to silicon treatment also led to the accumulation of analytes with osmoprotectant properties. For example, mucic acid, also known as galactaric acid, is a sugar acid, and acts as an osmoprotectant under drought stress conditions and can be used as a metabolite marker for stress tolerance.⁸⁵ In addition, methylated derivatives of myo-inositol, bornesitol and L-pinitol, which act as osmoregulators and as free radical scavengers were also accumulated.^86,87 These osmolytes were significantly accumulated in the roots of salinity-stressed plants.

The plant membrane glycoglycerolipid monogalactosyldiacylglycerol (MGDG 36:5)⁶⁸ was accumulated in the silicon-treated roots under saline conditions. This metabolite is an essential component of chloroplasts membranes. Overexpression of a rice monogalactosyldiacylglycerol synthase (OsMGD) gene, a key enzyme in the biosynthesis of MGDG, improved the salt tolerance of transgenic tobacco.⁸⁸

The observed differential accumulation of metabolites provided an insight into the presence of some detoxification mechanisms triggered by silicon. For example, the significant (p, q ≤ 0.05) accumulation of the detoxification intermediate S-formylglutathione in the roots of silicon-treated plants may suggest that silicon enhances the degradation of formaldehyde in plants, catalyzed by alcohol dehydrogenase-3 in the presence of reduced glutathione (GSH).⁸⁹ In the absence of silicon, there was a significant (p, q ≤ 0.05) increase in the accumulation of the cyanide detoxification intermediates beta-cyano-L-alanine and L-3-cyanoalanine, in the salinity-treated roots. These metabolites are involved in cyanide detoxification during ethylene biosynthesis.⁹⁰ This observation may suggest that there is an upregulation of the synthesis of ethylene under salinity treatment. However, in the presence of silicon, these metabolites were lower, probably because ethylene production was inhibited by silicon, since it was previously reported that silicon decreases the accumulation of 1-aminocyclopropane-1-carboxylic acid (ACC), the main precursor for the biosynthesis of ethylene in plants.⁹¹

The accumulation of S-D-lactoylglutathione in the silicon-treated plants under salinity suggests the presence of a detoxification mechanism for the cytotoxic MG, a byproduct of various metabolic processes.⁶⁸ Previous studies showed that MG acts as a signaling molecule regulating various processes of growth and development via the regulation of Ca⁺, K⁺, ROS and ABA.^92,93 However, hyperaccumulation of MG in the cell is detrimental, leading to the generation of ROS and the formation of advanced glycation end products.⁹³ To overcome the cytotoxicity of MG, plants are equipped with a glyoxalase detoxification system to maintain MG homeostasis. MG is detoxified via two main pathways: the glutathione-dependent pathway which is catalyzed via the two enzymes glyoxalase I (lactoylglutathione lyase) and glyoxalase II (hydroxyacylgluthathione hydrolase) and the glutathione-independent pathway which is catalyzed by glyoxalase III (DJ-1)⁶³ (Figure 9). The overaccumulation of the MG degradation intermediate S-D-lactoylgluthione in the silicon-treated date palm seedlings under salt stress conditions implies that the efficiency of the glyoxalase system in these plants is greater, compared with the control plants. Therefore, silicon treatment possibly enhances the rate of photosynthesis (A) and the root biomass (Figures 1 and 2).

The accumulation of di- and tri-amino acid molecules in the silicon-treated plants under salinity suggests the degradation and recycling of the oxidized-glutathione and glutathione-bound toxic metabolites via the gamma-glutamyl-transferase (GGT) cycle and cys-gly dipeptide.⁴⁶ Glutathione is an intracellular thiol molecule which is necessary for the plant growth and development.⁹⁴ It is a strong non-enzymatic antioxidant which plays an important role in protecting proteins from oxidative denaturation by protecting their thiol molecules under a variety of stress conditions. Therefore, the degradation and recycling of the oxidized-glutathione and glutathione-bound toxic metabolites are important, to maintain the cellular redox balance that may be challenged by various stresses. There is another potential alternative pathway, in which glutathione degrades in the cytosol, via gamma-glutamyl-cotransferase (GGCT) and 5-oxo-prolinase (5-Opase).³⁹ These two pathways exist side by side and are independent in function. The gamma-glutamyl cycle is important for extracellular glutathione recycling, whereas gamma-glutamyl-cotransferase (GGCT) and 5-oxo-prolinase (5-Opase) maintain the cytosolic glutathione homeostasis (Figure 10).

It was noted that some microbial and antimicrobial metabolites were identified during the LC-MS analysis. This is not a surprise because these metabolites may be derived from the endophytic microbes colonizing date palm tissues.

5. Conclusions

The use of silicon significantly enhances some of the growth parameters along with Casparian band formation in date palm seedlings. In general, metabolic profiles revealed the enhanced accumulation of stress-induced metabolites such as antioxidants and ROS scavengers, osmolytes, detoxified cytotoxic intermediates in a silicon-dependent manner in plants grown under saline conditions. Further studies are required to functionally characterize these differentially accumulated metabolites and to verify the mechanism by which silicon enhances salt tolerance in date palm.

Availability of datasets

LC-MS processed results obtained from this study were deposited in the Metabolomics Workbench database with the accession number ST001155.

Disclosure of potential conflicts of interest

The authors declare no conflict of interest.

Missing Value

Supplementary materials

Supplemental data for this article can be accessed on the publisher’s website.

Additional information

Funding

This study is supported by The Research Council of Oman (TRC), grant number 151.